SlideShare a Scribd company logo
Innovation Lab,
Nomura Asset Management Co., Ltd.
Kei Nakagawa
Copy Right© 2020 Kei Nakagawa
Contents
1. Introduction
2. Research Examples
- Time series prediction
- Cross sectional prediction
3. Conclusion
Contents
1. Introduction
2. Research Examples
- Time series prediction
- Cross sectional prediction
3. Conclusion
What is AI/Machine Learning?
・Cerebral Cortex:Unsupervised Learning
・Basal Ganglia:Reinforcement Learning
・Cerebellum:Supervised Learning
Machine Learning is one of the ways we expect to achieve AI.
- With the recent development of ML, ML is used in the same way as AI.
ML is an algorithm that imitates the function of the brain.
Learning means finding a model that represents the data well.
What is AI/Machine Learning?
𝑋:Feature
???Model:CNN
𝑌:Label
In case of image recognition, the goal is to identify the image
… Problem
Formulation
… Data
𝑓
In supervised learning, output is called label and input is called feature.
Supervised learning learn the relationship between inputs and outputs.
Cat
Deciding what model, features, and labels to use is called
problem formulation.
What is AI/Machine Learning?
Model:RNN
In case of translation, the goal is to translate English into Japanese
… Problem
Formulation
… Data
This is a pen.
???
This is a pen in Japanese.
𝑓
In supervised learning, output is called label and input is called feature.
Supervised learning learn the relationship between inputs and outputs.
𝑋:Feature 𝑌:Label
これはペンです
Deciding what model, features, and labels to use is called
problem formulation.
What is AI/Machine Learning?
Model:ML
In case of finance, the goal is to predict future stock prices.
… Problem
Formulation
… Data
???
+10%𝑓
In supervised learning, output is called label and input is called feature.
Supervised learning learn the relationship between inputs and outputs.
𝑋:Feature 𝑌:Label
Deciding what model, features, and labels to use is called
problem formulation.
A. Firstly, we need to decide what data we use and set up an
appropriate problem formulation.
Problem Formulation
• Time Series
• Cross Section
Data
• Technical
• Fundamental
• Alternative
AI/Machine Learning
How to predict stock prices w/ AI and ML?
Problem Formulation
クロスセクションで予測
Stock A Stock B ・・・
Today
Yesterday
・
・
・
・
Buystockswithabsolutely
highreturns. Buy stocks with relatively high returns.
Investment Universe
TimeSeries
Cross SectionOrthogonal
Problem Formulation:Time Series
Past (known) data at time t
Predict a stock future price/return as time series data.
・ 𝑟𝑡+1: (Absolute) return at time t+1 (Label)
・ 𝑟𝑡+1 = 𝑓(𝑟𝑡, … , 𝑥 𝑡, … )
・ 𝑟𝑡, 𝑥 𝑡: Past returns and external variables (Feature)
Short term forecast (shorter than monthly) → trading strategy
Problem Formulation: Learn the following model 𝑓
Problem Formulation:Cross Section
Predict multiple stocks that are relatively better in a universe.
・ 𝑠𝑐𝑜𝑟𝑒𝑖: (Relative) return of each stock in the universe (Label)
・ 𝑠𝑐𝑜𝑟𝑒𝑖 = 𝑓(𝑿𝒊)
・ 𝑿𝒊: Criteria for describing the score: Factor (Feature)
Long term forecast (longer than monthly) → portfolio construction
Time information is omitted
Problem Formulation: Learn the following model 𝑓
Investment
Horizon
Minutes Hour Day Week Month Year
Problem
Formulation
Time Series
Cross Section
Data Technical
Fundamental
Alternative Micro(Corporate)→Macro
News、PoS、etc,…
AI/Machine Learning
Problem Formulation:Summary
Contents
1. Introduction
2. Research Examples
- Time series prediction
- Cross sectional prediction
3. Conclusion
Time Series × Technical × Supervised
“Stock Price Prediction with Fluctuation Patterns Using Indexing
Dynamic Time Warping and k∗-Nearest Neighbors”.
Nakagawa et.al. (2018)
New Frontiers in Artificial Intelligence
https://doi.org/10.1007/978-3-319-93794-6_7
Motivation
Following nearest trend Following similar trend
Momentum Patterns
Predict the future price using the past price fluctuation patterns.
Refine the “formation analysis” by machine learning.
Nakagawa et.al. (2018) “Stock Price Prediction with Fluctuation Patterns Using
Indexing Dynamic Time Warping and k∗-Nearest Neighbors”.
New Frontiers in Artificial Intelligence
https://doi.org/10.1007/978-3-319-93794-6_7
Method - Outline
Current pattern
Umm...
Similar!!
1. Measure 2. Predict
・How to measure the similarity?
Indexing Dynamic Time Warping(IDTW)
・How to predict with extracted fluctuations?
k*-Nearest Neighbors (k*-NN)
We propose a method to predict future stock prices based on
the shape of the past price fluctuations similar to the current.
frequencymeasuring period
1. Decide the measuring period and frequency.
2. Consider the difference of price levels.
Near?
Distance
3. Define the distance between time-series.
✔ Daily closing prices within a month.
✔ Dynamic Time Warping (DTW) method
How to measure the similarity between stock price movements?
✔ We index prices and adjust price levels.
Method - IDTW
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
Correspondence
With Euclidean Distance
1
n
n
Sequences are aligned “one to one”.
●
●
● ● ●
●
●
●
●
●
● ●
●
●
●
●
●
● ●
Correspondence
With DTW
1
m
n
Nonlinear alignments are possible.
Fixed Time Axis “Warped” Time Axis
DTW is a method that warps the time axis to find the optimum
correspondence and calculates the similarity between two-time series.
What is DTW??
Why do we use DTW??
・ DTW can be applied to a pair of time-series of different lengths.
・ Similarity of DTW fits human intuition.
・ DTW is a standard method to measure the similarity among
various time series data (UCR Datasets)
What is DTW??
https://cran.r-project.org/web/packages/dtw/index.html
R language package dtw
Method - Outline
Current pattern
Umm...
Similar!!
1. Measure 2. Predict
・How to measure the similarity?
Indexing Dynamic Time Warping(IDTW)
・How to predict with extracted fluctuations?
k*-Nearest Neighbors (k*-NN)
We propose a method to predict future stock prices based on
past price fluctuations similar to the current.
Method - k-NN
✔ 𝑘-NN algorithm is one of the most fundamental non-parametric methods.
Similarity
𝑘 = 1 𝑘 = 3
Prediction:
+2.5%
Prediction:
(𝛼12.5 + 𝛼21.1 − 𝛼35.1)%
Return on
Next month
(Labels)
…
+2.5%
𝑦1
+1.1%
𝑦2
−5.1%
𝑦3
−3.2%
𝑦4
0.15𝛽1 0.24𝛽2 0.41𝛽3 0.63𝛽4 𝛽𝑖 = 𝐼𝐷𝑇𝑊 𝑥0, 𝑥𝑖
Fluctuation
pattern
(Features)
𝑥0
𝑥1 𝑥2 𝑥3 𝑥4
?%
✔ Prediction is the average of the values of 𝑘 nearest neighbors.
Method - k*-NN
✔ 𝑘-NN algorithm is one of the most fundamental non-parametric methods.
𝑘 = 1 𝑘 = 3
Prediction:
+2.5%
Prediction:
(𝛼12.5 + 𝛼21.1 − 𝛼35.1)%
…
+2.5%
𝑦1
+1.1%
𝑦2
−5.1%
𝑦3
−3.2%
𝑦4
?%
✔ Prediction is the average of the values of 𝑘 nearest neighbors.
1. How many k should we use to predict?
2. How to decide the optimal weight alpha?
𝑘∗
-NN is a simple approach for calculating optimal 𝒌 and weight 𝜶.
https://cran.r-project.org/web/packages/ksNN/index.html
R language package ksNN
Empirical Study
✔ To demonstrate the advantages of IDTW+k*NN,
we analyze its performance using following major world stock indices.
✔ In-sample period : January 1989 ~ December 2005.
Out-of-sample period : January 2006 ~ August 2017.
✔ Compared six methods combining DTW, DDTW, or IDTW with k-NN or k*-NN.
Ticker Name Country Constituents
TPX TOPIX Japan Companies listed on the First Section of the Tokyo Stock Exchange
SPX S&P500 U.S. 500 large companies listed on the NYSE or NASDAQ
CAC CAC40 France 40 companies listed on the Euronext Paris
DAX DAX German 30 companies trading on the Frankfurt Stock Exchange
UKX FTSE100 U.K. 100 companies listed on the London Stock Exchange
Empirical Study
Step 1. Calculate the similarity between month 𝑡 and all months from 1 to 𝑡 − 1
with DTW, DDTW, or IDTW.
𝑡
1 𝑡 − 1
Calculate each similarity
t-3 t-2 t-1 t21
DTW
DDTW
or
IDTW
or
Empirical Study
Step 2. Predict the next month's return using k-NN or k*-NN.
6
120
t-2
27
11
Sorting
by distance
+1.5%
0.02
0.03
0.04
7.99
8.82
+1.9%
-0.5%
+3.0%
-2.7%
𝛽
𝑦Next month’s
returnsdistances
k-NN
k*-NN
Predict
or
+𝟐. 𝟔%
Empirical Study
Step 3. If the predicted next month's return is positive, buy one unit of Index at
the end of the month. If negative, sell one unit and calculate the return.
Step 4. Proceed to the next month: 𝑡 = 𝑡 + 1.
𝑡
t t+1
𝑡 + 1
Index Price
(Month End)
100 105
Predict
+𝟐. 𝟔%
+𝟓. 𝟎%Actual
Results
✔ Cumulative returns for the six methods in TOPIX.
The out-of-sample period is from January 2006 to August 2017.
Red Line:Our method
(IDTW+k*NN)
Blue Line:
Buy & Hold Index
Results
Results
Table 1. The average accuracy of all years for each method.
Table 2. The total return of each method.
Results
Table 3. Comparison with Momentum Strategy(Return)
Table 4. Correlation with Momentum Strategy
Here, Prediction in Step 2 is replaced by simple 1Mom: 1 month return and
12 Mom: 12 month return excluding the last month (12 Mom).
Summary
✔ We proposed k*-NN with IDTW, a highly accurate stock prediction
method. This method is inspired by formation analysis.
✔ An empirical analysis was conducted with major world indices and
confirmed the following results:
・ IDTW is superior to DTW and DDTW
・ k*-NN is superior to k-NN.
・ IDTW-k*NN is the best prediction method.
Time Series × Technical × Unsupervised
“Stock price prediction using k‐medoids clustering with
indexing dynamic time warping”.
Nakagawa et.al. (2019)
Electronics and Communications in Japan, 102(2), 3-8.
https://doi.org/10.1002/ecj.12140
Motivation
Price movements
that are likely to rise?
✔ Can we visualize good patterns for forecasting by clustering price
fluctuation patterns?
Price movements
that are likely to fall?
✔ So how should we cluster the price fluctuation patterns?
Nakagawa et.al. (2019). “Stock price prediction using k‐medoids clustering with
indexing dynamic time warping”.
Electronics and Communications in Japan, 102(2), 3-8.
・ Always looks for medoids in the data (Needs only similarity).
✔ Since the similarity was given by IDTW, we should decide the clustering
method.
k-means method
k-medoids method
How to Cluster Price Fluctuation?
・ Most well-used clustering method
・ Clustering data based on centroid
・ Easy to extent (x-means, k-means++)
・ Clustering data based on medoids
(Data point closest to any data point in the cluster)
2007/1
2006/12
Clustering with k-Medoids
2004/5
Step1: Initialize medoids
(𝑘: number of medoids)
✔ Randomly select 𝑘 = 3 medoids in the data
2007/1
2006/12
2004/5
IDTW
Step2: Clustering
✔ Assign each data point to the cluster
to which the closest medoid belong
Step1: Initialize medoids
(𝑘: number of medoids)
Clustering with k-Medoids
2007/1
2006/12
Step3: Update medoids
✔ recalculate and update
medoids in each cluster
2004/5
Step2: ClusteringStep1: Initialize medoids
(𝑘: number of medoids) Repeat until the clusters
do not change
Clustering with k-Medoids
Empirical Study
✔ Using the TOPIX, we cluster similar price fluctuation patterns and visualize
good patterns for forecasting.
✔ In-sample period : January 1989 ~ December 2005.
Out-of-sample period : January 2007 ~ March 2017.
✔ IDTW with k-medoids is compared to DTW with k-medoids
✔ The number of k are from 2 to 12.
To decide k, we evaluate profitability of the following virtual trading.
k-medoids is performed based on
monthly daily price fluctuations up to
the previous month.
Identify the cluster for
the current month.
If there is more positive
(negative) in the cluster,
buy (sell).
Step 2:
Identify the cluster
Step 3:
Decide buy or sell
Step1:
Clustering
Empirical Study
✔ From the viewpoint of profitability, the number of k is 𝟓
Table 5. Profitability and accuracy
by number of clusters
Results
✔ Visualization of 5 clusters (black line: Medoids, x axis: days, y axis: return)
Reversal Momentum Momentum
Results
*The percentage in parentheses indicates the probability that stock price rises in the next month.
*
FlatReversal?
Results
✔ Visualization of 5 clusters (black line: Medoids, x axis: days, y axis: return)
✔ Using k-Medoids + IDTW, we found that the number of clusters in
the price fluctuation pattern was about 5 in TOPIX.
✔ Consider price fluctuation pattern clustering method (k-Medoids +
IDTW). This is unsupervised approach.
Summary
✔ Large fluctuations are tend to continue.
Time Series × Technical × Supervised
“Time-Series Gradient Boosting Tree for Stock Price Prediction”.
Nakagawa and Yoshida (2020)
International Journal of Data Mining, Modelling and Management
(IJDMMM)
To be appear
Fundamental
Motivation
✔ We improve a time series decision tree (TSDT) that uses only time
series data.
Nakagawa and Yoshida (2020)
“Time Series Gradient Booting Tree for Stock Price Prediction”.
International Journal of Data Mining, Modelling and Management (To be appear)
(Time series) decision trees
classify the (time series) data
by sorting them down the tree
and leaves provides the labels.
Decision tree is
easy to understand!
Current
Motivation
We propose a time-series gradient boosting tree that uses a TSDT
that adds features other than time-series.
(1) Add usual features other
than time series to TSDT
(2) “Boost” this TSDT with
gradient boosting technique
Method – Time Series Gradient Boosting Tree
✔ Boosting is one of the methods called ensemble learning.
✔ Gradient boosting tree is most used in various data analysis contests (Kaggle)
and has the best results.
strong model
weak model
Ensemble learning is a method of building a strong (accurate) model by
using multiple models called weak learners.
Method – Time Series Gradient Boosting Tree
Fit a new tree to the error
Prettenhofer and Louppe(2014)
Illustration of gradient boosting tree
Empirical Study
✔ To demonstrate the advantages of time series gradient boosting tree,
we analyze its performance using following major world stock indices.
✔ In-sample period : 10 years.
Out-of-sample period : 10 years later until 2018/6
Ticker Name Country Data period
TPX TOPIX Japan 1993/8~2018/6
SPX S&P500 U.S. 1990/1~2018/6
CAC CAC40 France 2001/6~2018/6
DAX DAX German 2001/7~2018/6
UKX FTSE100 U.K. 2000/1~2018/6
Empirical Study
✔ Feature:
Other Data :As of the end of month in each index, we use
✔ Model: Compared four methods combining Time Series Decision Tree
(TSDT) and Time Series Gradient Boosting Tree (TSGBT) with and
without index data.
Data Description
PER Net Profit/Market Value
PBR Net Asset/Market Value
MOM Stock returns over the past 12 months except for past month
DIV Dividends paid in a year / Market Value
ROE Net income/Net Assets
Time Series Data:Daily price data within a month
✔ Teacher:1 month ahead return
Results
Table 6. The average accuracy of all years for each method.
Table 7. The total return for each method.
Improvement
by adding features
Improvement
by boosting
Summary
✔ We build a time series decision tree (TSDT) by introducing features
other than time series data.
・Adding features other than time series data improves accuracy and profitability.
・Gradient boosting also improves accuracy and profitability
✔ Furthermore, we propose a time series gradient boosting tree
(TSGBT) that “boosts” time-series decision tree.
✔ An empirical analysis was conducted with major
world indices and confirmed the following results:
1. Introduction
2. Research Examples
- Time series prediction
- Cross sectional prediction
3. Conclusion
Contents
Cross Section × Technical × Supervised
“Deep Factor Model”.
Nakagawa et.al. (2018) 3rd Workshop on MIDAS@ECML-PKDD 2018
([preprint] arXiv:1810.01278)
“Cross-sectional Stock Price Prediction using Deep Learning for
Actual Investment Management”
Abe and Nakagawa (2020) AIBC2020
([preprint] :arXiv:2002.06975)
“Deep Learning for Multi-factor Models in Regional and Global Stock
Markets”
Abe and Nakagawa (2020) New Frontier in AI
https://link.springer.com/chapter/10.1007/978-3-030-58790-1_6
Fundamental
Cross Sectional Prediction
✔ Over 300 factors until 2012 [Harvey et al. 2017]
✔ In practice, we predict future relative stock returns (scores) by
combining various factors
✔ Criteria for describing the score is called Factor in Finance
For example, Market Value is one of the most famous factors, and this factor combined with
simple sorting portfolio yields a positive return.
Cross-sectional Investment Strategy
✔ finds stocks with relatively high scores based on certain criteria.
ROE
1 Month Return
●
●
●
Score
Value
Growth
Quality
Momentum
・Linear Relationship
Factor
Candidates
Factor Classification
by human
Calculate
Relative Score
Cross-sectional Investment Strategy
Both buying and selling are
evaluated equally.
Factors are evaluated monotonically.
ROE
1 Month Return
●
●
●
Score
Value
Growth
Quality
Momentum
・Linear Regression
Factor
Candidates
Factor Classification
by human
Calculate
Relative Score
Cross-sectional Investment Strategy
Both buying and selling are
evaluated equally.
Factors are evaluated monotonically.
I would like to capture the non-
linear relationship of factors from a
large number of candidates by
deep learning.
Cross-sectional Investment Strategy w/ DL
ROE
1 Month Return
●
●
●
Factor
Candidates
・
・
・
・
・
・
・
・
・
・
・
・
・
・
・
Extraction of effective factors in deep
learning (without human judgment)
Score
・Non-Linear Relationship
「Deep Factor Model」
Kei Nakagawa, Takumi Uchida and Tomohisa Aoshima,
3rd Workshop on MIDAS@ECML-PKDD 2018 ([preprint] arXiv:1810.01278)
✔ We check the profitability of cross-sectional investment strategy
with deep learning
✔ We predict relative return using 80 factors and invest monthly.
✔ Investment Universe:
TOPIX index constituents (from Apr 2006 to Mar 2016 monthly)
Empirical Study No.1
✔ We use the 16 factors that are used relatively often in practice.
Empirical Study No.1
Factor Candidates Formula
Risk
60VOL Standard deviation of stock returns in the past 60 months
BETA Regression coefficient of stock returns and market risk premium
SKEW Skewness of stock returns in the past 60 months
Quality
ROE Net income/Net Assets
ROA Operating Profit/Total Assets
ACCRUALS Operating Cashflow ‐ Operating Profit
LEVERAGE Total Liabilities / Total Assets
Momentum
12-1MOM Stock returns in the past 12 months except for past month
1MOM Stock returns in the past month
60MOM Stock returns in the past 60 months
Value
PSR Sales/Market Value
PER Net Income / Market Value
PBR Net Assets / Market Value
PCFR Operating Cashflow / Market Value
Size
CAP log(Market Value)
ILLIQ average(Stock Returns/Trading Volume)
✔ Use 16 x 5 = 80 features for the last 5 points each quarter of 16 factors.
Empirical Study No.1
✔ We Learn the model using the training data for 120 sets.
✔ Training data (1 set)
・Feature: 80 features of 16 factors
・Teacher: One month ahead relative returns
Model Description
Deep Learning(Deep 1) We use full-connected network. The hidden layers are (80-50-
10),(80-80-50-50-10-10). We use the ReLU as the activation
function and Adam for the optimization algorithm.Deep Learning(Deep 2)
Linear Model
(Linear Regression;LR)
“sklearn.linear_model.LinearRegression”
Support Vector
Regression(SVR)
“sklearn.svm.SVR”
Random Forest(RF) “sklearn.ensemble.RandomForestRegressor”
✔ We use linear and nonlinear models as benchmarks.
✔ We make a long/short portfolio strategy for a net-zero investment to buy
top stocks and to sell bottom stocks with equal weighting in quintile portfolios.
Empirical Study No.1
✔ For the quintile portfolio performance, we calculate the
annualized average return, and Sharpe ratio.
Results No.1
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
Deep 1 Deep 2 LR SVR RF
Return✔ This graph shows the (annualized) returns of each methods
The results of deep learning are better
than other nonlinear models.
The linear model is not bad either.
Results No.1
✔ This graph shows the Sharpe Ratio of each methods
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
Deep 1 Deep 2 LR SVR RF
ShapeRatio
The results of deep learning are better
than other nonlinear models.
The linear model is not bad either.
Linear model is good if a poor
nonlinear model is used.
Empirical Study No.2
✔ We check the profitability of cross-sectional investment
strategy with deep learning
✔ We predict relative return using 33 factors and invest daily basis.
✔ Investment Universe:
Cross-sectional Stock Price Prediction using Deep Learning for Actual Investment Management
Masaya Abe and Kei Nakagawa,
AIBC2020 ([preprint] arXiv: 2002.06975)
TOPIX500 index constituents (from Apr 2013 to Mar 2017 daily)
Empirical Study No.2
✔ We use the 33 factors that are used relatively often in practice.
No. Factor
1 Return from previous day
2 Return from 2 days ago
3 Return from 3 days ago
4 Return from 5 days ago
5 Return from 10 days ago
6 Return from 20 days ago
7 Return from 40 days ago
8 Return from 60 days ago
9 Average trading volume over the past 60 days
10 Average trading volume over the past 5 days/60 days
11 Average trading volume over the past 10 days/60 days
12 Average trading volume over the past 20 days/60 days
13 Change in operating income forecast from 5 days ago
14 Change in operating income forecast from 10 days ago
15 Change in operating income forecast from 20 days ago
16 Change in target stock price forecast from 5 days ago
17 Change in target stock price forecast from 10 days ago
No. Factor
18 Change in target stock price forecast from 20 days ago
19 Book-value to Price Ratio
20 Earnings to Price Ratio
21 Dividend Yield
22 Sales to Price Ratio
23 Cashflow to Price Ratio
24 Return on Equity
25 Return on Asset
26 Return on Invested Capital
27 Accruals
28 Total Asset Turnover Rate
29 Current Ratio
30 Equity Ratio
31 Total Asset Growth Rate
32 Capital Expenditure Growth Rate
33 Investment to Asset
- -
✔ We use linear and nonlinear models as benchmarks.
Empirical Study No.2
Model Hidden Layer (Dropout) Epoch
DNN_1
500 – 200 – 100 – 50 – 10
(50% – 40% – 30% – 20% – 10%)
20
DNN_2
500 – 200 – 100 – 50 – 10
(50% – 40% – 30% – 20% – 10%)
30
DNN_3
200 – 200 – 100 – 100 – 50
(50% – 50% – 30% –30% – 10%)
20
DNN_4
200 – 200 – 100 – 100 – 50
(50% – 50% – 30% – 30% – 10%)
30
DNN_5
300 – 300 – 150 – 150 – 50
(50% – 50% – 30% – 30% – 10%)
20
DNN_6
300 – 300 – 150 – 150 – 50
(50% – 50% – 30% – 30% – 10%)
30
・ Deep Learning ・ Random Forest (RF)
Model # of feature # of tree depth
RF_1 11 1000 3
RF_2 11 1000 5
RF_3 11 1000 7
model regularization
RR_1 0.1
RR_2 1.0
RR_3 10.0
・ Linear Model (Ridge Regression;RR)
✔ We make a quintile long/short portfolio and
calculate the annualized average return, and Sharpe ratio.
For the epoch number, 20 is good.
In this study, 30 is overfitting.
Empirical Study No.2
✔ This graph shows the (annualized) returns of each methods
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
Return
For the level of return,
the linear models(RR) are very good.
Empirical Study No.2
✔ This graph shows the Sharpe Ratio of each methods
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
ShapeRatio
For the epoch number, 20 is good.
In this study, 30 is overfitting.
For the Shape ratio,
DNNs are the very good.
Empirical Study No.3
✔ We check the profitability of cross-sectional investment
strategy with deep learning
✔ We predict relative return using 20 factors and invest monthly.
✔ Investment Universe:
Deep Learning for Multi-factor Models in Regional and Global Stock Markets
Masaya Abe and Kei Nakagawa,
New Frontiers in Artificial Intelligence (To be appear)
World : MSCI WORLD
North America: MSCI North Americas
Europe: MSCI Europe & Middle East
Asia Pacific: MSCI Pacific
(from 2005 to 2017 monthly)
Empirical Study No.3
✔ We use the 20 factors that are used relatively often in practice.
No. Factor No. Factor No. Factor
1 Book-to-market ratio 8 Return on invested capital 15 EPS Revision(1 month)
2 Earnings-to-price ratio 9 Accruals 16 EPS Revision(3 months)
3 Dividend yield 10 Total Asset Growth Rate 17 Past stock return(1 month)
4 Sales-to-price ratio 11 Current ratio 18 Past stock return(12 months)
5 Cash flow-to-price ratio 12 Equity ratio 19 Volatility
6 Return on equity 13 Total Asset Turnover Rate 20 Skewness
7 Return on asset 14 CAPEX Growth Rate
✔ We use linear and nonlinear models as benchmarks.
Empirical Study No.3
Model Description
Deep Learning
We use full-connected network. The hidden layers are (150-
150-100-100-50-50). Dropout rate: (50% – 50% – 30% –
30% – 10% – 10%).We use the ReLU as the activation function
and Adam for the optimization algorithm.
Linear Model
(Ridge Regression)
“sklearn.linear_model.Ridge”
Support Vector Regression “sklearn.svm.SVR”
Random Forest “sklearn.ensemble.RandomForestRegressor”
Gradient Boosting Tree “XGBRegressor”
✔ We make a quintile long/short portfolio and
calculate the annualized average return, and Sharpe ratio.
Results No.3
✔ This graph shows the (annualized) returns of each methods
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
DNN
RR
XGB
RF
DNN
RR
XGB
RF
DNN
RR
XGB
RF
DNN
RR
XGB
RF
NA EU PF WLD
Return
✓ DNN generally outperforms other
methods in all regions.
✓ The performance decreases in the
order of PF, EU, NA regardless of
the prediction model.
Results No.3
✔ This graph shows the Sharpe Ratio of each methods
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40 DNN
RR
XGB
RF
DNN
RR
XGB
RF
DNN
RR
XGB
RF
DNN
RR
XGB
RF
NA EU PF WLD
ShapeRatio
✓ The highest Shape ratios are given
by DNN in all regions.
✓ The performance decreases in the
order of PF, EU, NA regardless of
the prediction model.
Summary
DNN is generally superior to GB, RF, and RR in Japan and global markets
→ DNN shows promise as a skillful machine learning method in the cross section.
The performance deteriorates in the order of Pacific,
Europe, North America regardless of the prediction model.
→ The higher the market efficiency, the more difficult to earn returns
Contents
1. Introduction
2. Research Examples
- Time series prediction
- Cross sectional prediction
3. Conclusion
Problem Formulation
• Time Series
• Cross Section
Data
• Technical
• Fundamental
• Alternative
AI/Machine Learning
Conclusion
If you have any question or comment, please feel free to Email me
kei.nak.0315@gmail.com
Disclaimer
The Information in this presentation is provided for education and informational purposes only, without any
express or implied warranty of any kind, including warranties of accuracy, completeness, or fitness for any
particular purpose. The Information contained in or provided from or through this material is not intended to be
and does not constitute financial advice, investment advice, trading advice or any other advice. The Information
in this material is general in nature and is not specific to you the viewer or anyone else.
If this document contains any information relating to the past performance of any securities or other investment
strategies and products, you should note that past performance is not indicative of future results. Any market or
investment view mentioned in this document is not intended to be investment research. Information contained in
this document is subject to change without any obligation on our part to notify you of any change. If this
document forms part of a presentation or is presented to you together with other documents and materials, this
document should not be read in isolation and may not provide a full explanation of all of the topics presented and
discussed. You should note that investment involves risk. The price of any securities or other investment product
may go down as well as up. The information does not accept any liability or legal responsibility whatsoever for
any decisions taken based upon the material.
∗ 𝑑(𝑥, 𝑦) = 𝑥 − 𝑦
∗
79
▷ Indexing
▷ DTW
Appendix Algorithm of IDTW
80
Appendix Algorithm of k*NN
81
▷ Step 1
▷ Step 2
▷ Step 3
Appendix Algorithm of k-medoids with IDTW
82
▷ Time Series
▷ Data
Appendix Algorithm of TSDT with usual data
83
Appendix Algorithm of TSGBT with usual data

More Related Content

What's hot

セミパラメトリック推論の基礎
セミパラメトリック推論の基礎セミパラメトリック推論の基礎
セミパラメトリック推論の基礎
Daisuke Yoneoka
 
アセットアロケーションの未来
アセットアロケーションの未来アセットアロケーションの未来
アセットアロケーションの未来
Kei Nakagawa
 
Arima model (time series)
Arima model (time series)Arima model (time series)
Arima model (time series)
Kumar P
 
Time series Analysis
Time series AnalysisTime series Analysis
Time series Analysis
Mahak Vijayvargiya
 
観察データを用いた因果推論に共変量選択
観察データを用いた因果推論に共変量選択観察データを用いた因果推論に共変量選択
観察データを用いた因果推論に共変量選択
Jaehyun Song
 
Chap16 time series analysis and forecasting
Chap16 time series analysis and forecastingChap16 time series analysis and forecasting
Chap16 time series analysis and forecasting
Judianto Nugroho
 
金融リスクとポートフォリオマネジメント
金融リスクとポートフォリオマネジメント金融リスクとポートフォリオマネジメント
金融リスクとポートフォリオマネジメント
Kei Nakagawa
 
予測理論とpredictability
予測理論とpredictability予測理論とpredictability
予測理論とpredictability
logics-of-blue
 
Time series forecasting
Time series forecastingTime series forecasting
Time series forecasting
Firas Kastantin
 
Capm
CapmCapm
Risk based portfolio with large dynamic covariance matrices
Risk based portfolio with large dynamic covariance matricesRisk based portfolio with large dynamic covariance matrices
Risk based portfolio with large dynamic covariance matrices
Kei Nakagawa
 
ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張
ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張
ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張
Kei Nakagawa
 
LSTM (Long short-term memory) 概要
LSTM (Long short-term memory) 概要LSTM (Long short-term memory) 概要
LSTM (Long short-term memory) 概要
Kenji Urai
 
GEE(一般化推定方程式)の理論
GEE(一般化推定方程式)の理論GEE(一般化推定方程式)の理論
GEE(一般化推定方程式)の理論
Koichiro Gibo
 
ファクター投資と機械学習
ファクター投資と機械学習ファクター投資と機械学習
ファクター投資と機械学習
Kei Nakagawa
 
Risk-Aversion, Risk-Premium and Utility Theory
Risk-Aversion, Risk-Premium and Utility TheoryRisk-Aversion, Risk-Premium and Utility Theory
Risk-Aversion, Risk-Premium and Utility Theory
Ashwin Rao
 
Time series Forecasting
Time series ForecastingTime series Forecasting
Time series Forecasting
haroonrashidlone
 
Lesson 2 stationary_time_series
Lesson 2 stationary_time_seriesLesson 2 stationary_time_series
Lesson 2 stationary_time_series
ankit_ppt
 
ARIMA
ARIMA ARIMA
Lesson 5 arima
Lesson 5 arimaLesson 5 arima
Lesson 5 arima
ankit_ppt
 

What's hot (20)

セミパラメトリック推論の基礎
セミパラメトリック推論の基礎セミパラメトリック推論の基礎
セミパラメトリック推論の基礎
 
アセットアロケーションの未来
アセットアロケーションの未来アセットアロケーションの未来
アセットアロケーションの未来
 
Arima model (time series)
Arima model (time series)Arima model (time series)
Arima model (time series)
 
Time series Analysis
Time series AnalysisTime series Analysis
Time series Analysis
 
観察データを用いた因果推論に共変量選択
観察データを用いた因果推論に共変量選択観察データを用いた因果推論に共変量選択
観察データを用いた因果推論に共変量選択
 
Chap16 time series analysis and forecasting
Chap16 time series analysis and forecastingChap16 time series analysis and forecasting
Chap16 time series analysis and forecasting
 
金融リスクとポートフォリオマネジメント
金融リスクとポートフォリオマネジメント金融リスクとポートフォリオマネジメント
金融リスクとポートフォリオマネジメント
 
予測理論とpredictability
予測理論とpredictability予測理論とpredictability
予測理論とpredictability
 
Time series forecasting
Time series forecastingTime series forecasting
Time series forecasting
 
Capm
CapmCapm
Capm
 
Risk based portfolio with large dynamic covariance matrices
Risk based portfolio with large dynamic covariance matricesRisk based portfolio with large dynamic covariance matrices
Risk based portfolio with large dynamic covariance matrices
 
ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張
ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張
ブラック・リッターマン法を用いたリスクベース・ポートフォリオの拡張
 
LSTM (Long short-term memory) 概要
LSTM (Long short-term memory) 概要LSTM (Long short-term memory) 概要
LSTM (Long short-term memory) 概要
 
GEE(一般化推定方程式)の理論
GEE(一般化推定方程式)の理論GEE(一般化推定方程式)の理論
GEE(一般化推定方程式)の理論
 
ファクター投資と機械学習
ファクター投資と機械学習ファクター投資と機械学習
ファクター投資と機械学習
 
Risk-Aversion, Risk-Premium and Utility Theory
Risk-Aversion, Risk-Premium and Utility TheoryRisk-Aversion, Risk-Premium and Utility Theory
Risk-Aversion, Risk-Premium and Utility Theory
 
Time series Forecasting
Time series ForecastingTime series Forecasting
Time series Forecasting
 
Lesson 2 stationary_time_series
Lesson 2 stationary_time_seriesLesson 2 stationary_time_series
Lesson 2 stationary_time_series
 
ARIMA
ARIMA ARIMA
ARIMA
 
Lesson 5 arima
Lesson 5 arimaLesson 5 arima
Lesson 5 arima
 

Similar to IFTA2020 Kei Nakagawa

CS3114_09212011.ppt
CS3114_09212011.pptCS3114_09212011.ppt
CS3114_09212011.ppt
Arumugam90
 
RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...
RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...
RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...
Kei Nakagawa
 
Deep Learning Introduction - WeCloudData
Deep Learning Introduction - WeCloudDataDeep Learning Introduction - WeCloudData
Deep Learning Introduction - WeCloudData
WeCloudData
 
Machine learning ppt unit one syllabuspptx
Machine learning ppt unit one syllabuspptxMachine learning ppt unit one syllabuspptx
Machine learning ppt unit one syllabuspptx
VenkateswaraBabuRavi
 
IRJET- Stock Market Prediction using Machine Learning
IRJET- Stock Market Prediction using Machine LearningIRJET- Stock Market Prediction using Machine Learning
IRJET- Stock Market Prediction using Machine Learning
IRJET Journal
 
Stock market analysis using supervised machine learning
Stock market analysis using supervised machine learningStock market analysis using supervised machine learning
Stock market analysis using supervised machine learning
Priyanshu Gandhi
 
Forecasting
ForecastingForecasting
Forecasting
Vikash Kumar
 
Forecasting
ForecastingForecasting
Forecasting
Irfan Hussain
 
Application_of_Deep_Learning_Techniques.pptx
Application_of_Deep_Learning_Techniques.pptxApplication_of_Deep_Learning_Techniques.pptx
Application_of_Deep_Learning_Techniques.pptx
KiranKumar918931
 
Reinforcement learning
Reinforcement learningReinforcement learning
Reinforcement learning
DongHyun Kwak
 
Applying Deep Learning to Enhance Momentum Trading Strategies in Stocks
Applying Deep Learning to Enhance Momentum Trading Strategies in StocksApplying Deep Learning to Enhance Momentum Trading Strategies in Stocks
Applying Deep Learning to Enhance Momentum Trading Strategies in Stocks
Lawrence Takeuchi
 
Chapter-3_Heizer_S1.pptx
Chapter-3_Heizer_S1.pptxChapter-3_Heizer_S1.pptx
Chapter-3_Heizer_S1.pptx
EdwardDelaCruz14
 
Essentials of machine learning algorithms
Essentials of machine learning algorithmsEssentials of machine learning algorithms
Essentials of machine learning algorithms
Arunangsu Sahu
 
Predictive model based on Supervised ML
Predictive model based on Supervised MLPredictive model based on Supervised ML
Predictive model based on Supervised ML
UmeshchandraYadav5
 
Forecasting_Quantitative Forecasting.pptx
Forecasting_Quantitative Forecasting.pptxForecasting_Quantitative Forecasting.pptx
Forecasting_Quantitative Forecasting.pptx
RituparnaDas584083
 
Machine learning: Stock Price Prediction
Machine learning: Stock Price PredictionMachine learning: Stock Price Prediction
Machine learning: Stock Price Prediction
eurosigdoc acm
 
fINAL ML PPT.pptx
fINAL ML PPT.pptxfINAL ML PPT.pptx
fINAL ML PPT.pptx
19445KNithinbabu
 
1 forecasting SHORT NOTES FOR ESE AND GATE
1 forecasting SHORT NOTES FOR ESE AND GATE1 forecasting SHORT NOTES FOR ESE AND GATE
1 forecasting SHORT NOTES FOR ESE AND GATE
Aditya Pal
 

Similar to IFTA2020 Kei Nakagawa (20)

CS3114_09212011.ppt
CS3114_09212011.pptCS3114_09212011.ppt
CS3114_09212011.ppt
 
RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...
RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...
RIC-NN: A Robust Transferable Deep Learning Framework for Cross-sectional Inv...
 
Deep Learning Introduction - WeCloudData
Deep Learning Introduction - WeCloudDataDeep Learning Introduction - WeCloudData
Deep Learning Introduction - WeCloudData
 
Presentation1
Presentation1Presentation1
Presentation1
 
Machine learning ppt unit one syllabuspptx
Machine learning ppt unit one syllabuspptxMachine learning ppt unit one syllabuspptx
Machine learning ppt unit one syllabuspptx
 
IRJET- Stock Market Prediction using Machine Learning
IRJET- Stock Market Prediction using Machine LearningIRJET- Stock Market Prediction using Machine Learning
IRJET- Stock Market Prediction using Machine Learning
 
Stock market analysis using supervised machine learning
Stock market analysis using supervised machine learningStock market analysis using supervised machine learning
Stock market analysis using supervised machine learning
 
Forecasting
ForecastingForecasting
Forecasting
 
Forecasting
ForecastingForecasting
Forecasting
 
Application_of_Deep_Learning_Techniques.pptx
Application_of_Deep_Learning_Techniques.pptxApplication_of_Deep_Learning_Techniques.pptx
Application_of_Deep_Learning_Techniques.pptx
 
Reinforcement learning
Reinforcement learningReinforcement learning
Reinforcement learning
 
Applying Deep Learning to Enhance Momentum Trading Strategies in Stocks
Applying Deep Learning to Enhance Momentum Trading Strategies in StocksApplying Deep Learning to Enhance Momentum Trading Strategies in Stocks
Applying Deep Learning to Enhance Momentum Trading Strategies in Stocks
 
Chapter-3_Heizer_S1.pptx
Chapter-3_Heizer_S1.pptxChapter-3_Heizer_S1.pptx
Chapter-3_Heizer_S1.pptx
 
Essentials of machine learning algorithms
Essentials of machine learning algorithmsEssentials of machine learning algorithms
Essentials of machine learning algorithms
 
Predictive model based on Supervised ML
Predictive model based on Supervised MLPredictive model based on Supervised ML
Predictive model based on Supervised ML
 
forecasting methods
forecasting methodsforecasting methods
forecasting methods
 
Forecasting_Quantitative Forecasting.pptx
Forecasting_Quantitative Forecasting.pptxForecasting_Quantitative Forecasting.pptx
Forecasting_Quantitative Forecasting.pptx
 
Machine learning: Stock Price Prediction
Machine learning: Stock Price PredictionMachine learning: Stock Price Prediction
Machine learning: Stock Price Prediction
 
fINAL ML PPT.pptx
fINAL ML PPT.pptxfINAL ML PPT.pptx
fINAL ML PPT.pptx
 
1 forecasting SHORT NOTES FOR ESE AND GATE
1 forecasting SHORT NOTES FOR ESE AND GATE1 forecasting SHORT NOTES FOR ESE AND GATE
1 forecasting SHORT NOTES FOR ESE AND GATE
 

More from Kei Nakagawa

AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討
AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討
AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討
Kei Nakagawa
 
単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド
単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド
単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド
Kei Nakagawa
 
連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)
連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)
連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)
Kei Nakagawa
 
Sigfin Neural Fractional SDE NET
Sigfin Neural Fractional SDE NETSigfin Neural Fractional SDE NET
Sigfin Neural Fractional SDE NET
Kei Nakagawa
 
B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~
B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~
B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~
Kei Nakagawa
 
What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...
What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...
What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...
Kei Nakagawa
 
RIC-NN:深層転移学習を用いたマルチファクター運用
RIC-NN:深層転移学習を用いたマルチファクター運用RIC-NN:深層転移学習を用いたマルチファクター運用
RIC-NN:深層転移学習を用いたマルチファクター運用
Kei Nakagawa
 
因果チェーンを用いたリードラグ効果の実証分析
因果チェーンを用いたリードラグ効果の実証分析因果チェーンを用いたリードラグ効果の実証分析
因果チェーンを用いたリードラグ効果の実証分析
Kei Nakagawa
 
A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...
A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...
A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...
Kei Nakagawa
 
Economic Causal Chain and Predictable Stock Returns
Economic Causal Chain and Predictable Stock ReturnsEconomic Causal Chain and Predictable Stock Returns
Economic Causal Chain and Predictable Stock Returns
Kei Nakagawa
 
GARCHSKモデルを用いた条件付き固有モーメントの実証分析
GARCHSKモデルを用いた条件付き固有モーメントの実証分析GARCHSKモデルを用いた条件付き固有モーメントの実証分析
GARCHSKモデルを用いた条件付き固有モーメントの実証分析
Kei Nakagawa
 
リスクベースポートフォリオの高次モーメントへの拡張
リスクベースポートフォリオの高次モーメントへの拡張リスクベースポートフォリオの高次モーメントへの拡張
リスクベースポートフォリオの高次モーメントへの拡張
Kei Nakagawa
 

More from Kei Nakagawa (12)

AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討
AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討
AIトレーダーが市場へ与える影響 –GARCH型モデルのミクロ的基礎づけによる検討
 
単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド
単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド
単調回帰を用いた一般化トレンド・ファクター:暗号資産市場への応用 SigFin発表スライド
 
連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)
連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)
連続時間フラクショナル・トピックモデル(NLP2023 金融・経済ドメインのための言語処理)
 
Sigfin Neural Fractional SDE NET
Sigfin Neural Fractional SDE NETSigfin Neural Fractional SDE NET
Sigfin Neural Fractional SDE NET
 
B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~
B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~
B2B企業ブランド価値の財務指標・株式市場へのインパクト ~PBR(株価純資産倍率)等への影響~
 
What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...
What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...
What Do Good Integrated Reports Tell Us?: An Empirical Study of Japanese Comp...
 
RIC-NN:深層転移学習を用いたマルチファクター運用
RIC-NN:深層転移学習を用いたマルチファクター運用RIC-NN:深層転移学習を用いたマルチファクター運用
RIC-NN:深層転移学習を用いたマルチファクター運用
 
因果チェーンを用いたリードラグ効果の実証分析
因果チェーンを用いたリードラグ効果の実証分析因果チェーンを用いたリードラグ効果の実証分析
因果チェーンを用いたリードラグ効果の実証分析
 
A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...
A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...
A Robust Transferable Deep Learning Framework for Cross-sectional Investment ...
 
Economic Causal Chain and Predictable Stock Returns
Economic Causal Chain and Predictable Stock ReturnsEconomic Causal Chain and Predictable Stock Returns
Economic Causal Chain and Predictable Stock Returns
 
GARCHSKモデルを用いた条件付き固有モーメントの実証分析
GARCHSKモデルを用いた条件付き固有モーメントの実証分析GARCHSKモデルを用いた条件付き固有モーメントの実証分析
GARCHSKモデルを用いた条件付き固有モーメントの実証分析
 
リスクベースポートフォリオの高次モーメントへの拡張
リスクベースポートフォリオの高次モーメントへの拡張リスクベースポートフォリオの高次モーメントへの拡張
リスクベースポートフォリオの高次モーメントへの拡張
 

Recently uploaded

Tumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdf
Tumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdfTumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdf
Tumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdf
Henry Tapper
 
2. Elemental Economics - Mineral demand.pdf
2. Elemental Economics - Mineral demand.pdf2. Elemental Economics - Mineral demand.pdf
2. Elemental Economics - Mineral demand.pdf
Neal Brewster
 
Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...
Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...
Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...
Godwin Emmanuel Oyedokun MBA MSc PhD FCA FCTI FCNA CFE FFAR
 
Earn a passive income with prosocial investing
Earn a passive income with prosocial investingEarn a passive income with prosocial investing
Earn a passive income with prosocial investing
Colin R. Turner
 
The secret way to sell pi coins effortlessly.
The secret way to sell pi coins effortlessly.The secret way to sell pi coins effortlessly.
The secret way to sell pi coins effortlessly.
DOT TECH
 
Scope Of Macroeconomics introduction and basic theories
Scope Of Macroeconomics introduction and basic theoriesScope Of Macroeconomics introduction and basic theories
Scope Of Macroeconomics introduction and basic theories
nomankalyar153
 
5 Tips for Creating Standard Financial Reports
5 Tips for Creating Standard Financial Reports5 Tips for Creating Standard Financial Reports
5 Tips for Creating Standard Financial Reports
EasyReports
 
Tdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFi
Tdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFiTdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFi
Tdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFi
nimaruinazawa258
 
Which Crypto to Buy Today for Short-Term in May-June 2024.pdf
Which Crypto to Buy Today for Short-Term in May-June 2024.pdfWhich Crypto to Buy Today for Short-Term in May-June 2024.pdf
Which Crypto to Buy Today for Short-Term in May-June 2024.pdf
Kezex (KZX)
 
Intro_Economics_ GPresentation Week 4.pptx
Intro_Economics_ GPresentation Week 4.pptxIntro_Economics_ GPresentation Week 4.pptx
Intro_Economics_ GPresentation Week 4.pptx
shetivia
 
Instant Issue Debit Cards - High School Spirit
Instant Issue Debit Cards - High School SpiritInstant Issue Debit Cards - High School Spirit
Instant Issue Debit Cards - High School Spirit
egoetzinger
 
Globalization (Nike) Presentation PPT Poster Infographic.pdf
Globalization (Nike) Presentation PPT Poster Infographic.pdfGlobalization (Nike) Presentation PPT Poster Infographic.pdf
Globalization (Nike) Presentation PPT Poster Infographic.pdf
VohnArchieEdjan
 
NEW NORMAL! WHAT BECOMES OF ACCOUNTING PROFESSION
NEW NORMAL!  WHAT BECOMES OF ACCOUNTING PROFESSION NEW NORMAL!  WHAT BECOMES OF ACCOUNTING PROFESSION
NEW NORMAL! WHAT BECOMES OF ACCOUNTING PROFESSION
Godwin Emmanuel Oyedokun MBA MSc PhD FCA FCTI FCNA CFE FFAR
 
Turin Startup Ecosystem 2024 - Ricerca sulle Startup e il Sistema dell'Innov...
Turin Startup Ecosystem 2024  - Ricerca sulle Startup e il Sistema dell'Innov...Turin Startup Ecosystem 2024  - Ricerca sulle Startup e il Sistema dell'Innov...
Turin Startup Ecosystem 2024 - Ricerca sulle Startup e il Sistema dell'Innov...
Quotidiano Piemontese
 
APP I Lecture Notes to students 0f 4the year
APP I  Lecture Notes  to students 0f 4the yearAPP I  Lecture Notes  to students 0f 4the year
APP I Lecture Notes to students 0f 4the year
telilaalilemlem
 
when will pi network coin be available on crypto exchange.
when will pi network coin be available on crypto exchange.when will pi network coin be available on crypto exchange.
when will pi network coin be available on crypto exchange.
DOT TECH
 
can I really make money with pi network.
can I really make money with pi network.can I really make money with pi network.
can I really make money with pi network.
DOT TECH
 
GeM ppt in railway for presentation on gem
GeM ppt in railway  for presentation on gemGeM ppt in railway  for presentation on gem
GeM ppt in railway for presentation on gem
CwierAsn
 
What price will pi network be listed on exchanges
What price will pi network be listed on exchangesWhat price will pi network be listed on exchanges
What price will pi network be listed on exchanges
DOT TECH
 
Analyzing the instability of equilibrium in thr harrod domar model
Analyzing the instability of equilibrium in thr harrod domar modelAnalyzing the instability of equilibrium in thr harrod domar model
Analyzing the instability of equilibrium in thr harrod domar model
ManthanBhardwaj4
 

Recently uploaded (20)

Tumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdf
Tumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdfTumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdf
Tumelo-deep-dive-into-pass-through-voting-Feb23 (1).pdf
 
2. Elemental Economics - Mineral demand.pdf
2. Elemental Economics - Mineral demand.pdf2. Elemental Economics - Mineral demand.pdf
2. Elemental Economics - Mineral demand.pdf
 
Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...
Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...
Tax System, Behaviour, Justice, and Voluntary Compliance Culture in Nigeria -...
 
Earn a passive income with prosocial investing
Earn a passive income with prosocial investingEarn a passive income with prosocial investing
Earn a passive income with prosocial investing
 
The secret way to sell pi coins effortlessly.
The secret way to sell pi coins effortlessly.The secret way to sell pi coins effortlessly.
The secret way to sell pi coins effortlessly.
 
Scope Of Macroeconomics introduction and basic theories
Scope Of Macroeconomics introduction and basic theoriesScope Of Macroeconomics introduction and basic theories
Scope Of Macroeconomics introduction and basic theories
 
5 Tips for Creating Standard Financial Reports
5 Tips for Creating Standard Financial Reports5 Tips for Creating Standard Financial Reports
5 Tips for Creating Standard Financial Reports
 
Tdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFi
Tdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFiTdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFi
Tdasx: Unveiling the Trillion-Dollar Potential of Bitcoin DeFi
 
Which Crypto to Buy Today for Short-Term in May-June 2024.pdf
Which Crypto to Buy Today for Short-Term in May-June 2024.pdfWhich Crypto to Buy Today for Short-Term in May-June 2024.pdf
Which Crypto to Buy Today for Short-Term in May-June 2024.pdf
 
Intro_Economics_ GPresentation Week 4.pptx
Intro_Economics_ GPresentation Week 4.pptxIntro_Economics_ GPresentation Week 4.pptx
Intro_Economics_ GPresentation Week 4.pptx
 
Instant Issue Debit Cards - High School Spirit
Instant Issue Debit Cards - High School SpiritInstant Issue Debit Cards - High School Spirit
Instant Issue Debit Cards - High School Spirit
 
Globalization (Nike) Presentation PPT Poster Infographic.pdf
Globalization (Nike) Presentation PPT Poster Infographic.pdfGlobalization (Nike) Presentation PPT Poster Infographic.pdf
Globalization (Nike) Presentation PPT Poster Infographic.pdf
 
NEW NORMAL! WHAT BECOMES OF ACCOUNTING PROFESSION
NEW NORMAL!  WHAT BECOMES OF ACCOUNTING PROFESSION NEW NORMAL!  WHAT BECOMES OF ACCOUNTING PROFESSION
NEW NORMAL! WHAT BECOMES OF ACCOUNTING PROFESSION
 
Turin Startup Ecosystem 2024 - Ricerca sulle Startup e il Sistema dell'Innov...
Turin Startup Ecosystem 2024  - Ricerca sulle Startup e il Sistema dell'Innov...Turin Startup Ecosystem 2024  - Ricerca sulle Startup e il Sistema dell'Innov...
Turin Startup Ecosystem 2024 - Ricerca sulle Startup e il Sistema dell'Innov...
 
APP I Lecture Notes to students 0f 4the year
APP I  Lecture Notes  to students 0f 4the yearAPP I  Lecture Notes  to students 0f 4the year
APP I Lecture Notes to students 0f 4the year
 
when will pi network coin be available on crypto exchange.
when will pi network coin be available on crypto exchange.when will pi network coin be available on crypto exchange.
when will pi network coin be available on crypto exchange.
 
can I really make money with pi network.
can I really make money with pi network.can I really make money with pi network.
can I really make money with pi network.
 
GeM ppt in railway for presentation on gem
GeM ppt in railway  for presentation on gemGeM ppt in railway  for presentation on gem
GeM ppt in railway for presentation on gem
 
What price will pi network be listed on exchanges
What price will pi network be listed on exchangesWhat price will pi network be listed on exchanges
What price will pi network be listed on exchanges
 
Analyzing the instability of equilibrium in thr harrod domar model
Analyzing the instability of equilibrium in thr harrod domar modelAnalyzing the instability of equilibrium in thr harrod domar model
Analyzing the instability of equilibrium in thr harrod domar model
 

IFTA2020 Kei Nakagawa

  • 1. Innovation Lab, Nomura Asset Management Co., Ltd. Kei Nakagawa Copy Right© 2020 Kei Nakagawa
  • 2. Contents 1. Introduction 2. Research Examples - Time series prediction - Cross sectional prediction 3. Conclusion
  • 3. Contents 1. Introduction 2. Research Examples - Time series prediction - Cross sectional prediction 3. Conclusion
  • 4. What is AI/Machine Learning? ・Cerebral Cortex:Unsupervised Learning ・Basal Ganglia:Reinforcement Learning ・Cerebellum:Supervised Learning Machine Learning is one of the ways we expect to achieve AI. - With the recent development of ML, ML is used in the same way as AI. ML is an algorithm that imitates the function of the brain. Learning means finding a model that represents the data well.
  • 5. What is AI/Machine Learning? 𝑋:Feature ???Model:CNN 𝑌:Label In case of image recognition, the goal is to identify the image … Problem Formulation … Data 𝑓 In supervised learning, output is called label and input is called feature. Supervised learning learn the relationship between inputs and outputs. Cat Deciding what model, features, and labels to use is called problem formulation.
  • 6. What is AI/Machine Learning? Model:RNN In case of translation, the goal is to translate English into Japanese … Problem Formulation … Data This is a pen. ??? This is a pen in Japanese. 𝑓 In supervised learning, output is called label and input is called feature. Supervised learning learn the relationship between inputs and outputs. 𝑋:Feature 𝑌:Label これはペンです Deciding what model, features, and labels to use is called problem formulation.
  • 7. What is AI/Machine Learning? Model:ML In case of finance, the goal is to predict future stock prices. … Problem Formulation … Data ??? +10%𝑓 In supervised learning, output is called label and input is called feature. Supervised learning learn the relationship between inputs and outputs. 𝑋:Feature 𝑌:Label Deciding what model, features, and labels to use is called problem formulation.
  • 8. A. Firstly, we need to decide what data we use and set up an appropriate problem formulation. Problem Formulation • Time Series • Cross Section Data • Technical • Fundamental • Alternative AI/Machine Learning How to predict stock prices w/ AI and ML?
  • 9. Problem Formulation クロスセクションで予測 Stock A Stock B ・・・ Today Yesterday ・ ・ ・ ・ Buystockswithabsolutely highreturns. Buy stocks with relatively high returns. Investment Universe TimeSeries Cross SectionOrthogonal
  • 10. Problem Formulation:Time Series Past (known) data at time t Predict a stock future price/return as time series data. ・ 𝑟𝑡+1: (Absolute) return at time t+1 (Label) ・ 𝑟𝑡+1 = 𝑓(𝑟𝑡, … , 𝑥 𝑡, … ) ・ 𝑟𝑡, 𝑥 𝑡: Past returns and external variables (Feature) Short term forecast (shorter than monthly) → trading strategy Problem Formulation: Learn the following model 𝑓
  • 11. Problem Formulation:Cross Section Predict multiple stocks that are relatively better in a universe. ・ 𝑠𝑐𝑜𝑟𝑒𝑖: (Relative) return of each stock in the universe (Label) ・ 𝑠𝑐𝑜𝑟𝑒𝑖 = 𝑓(𝑿𝒊) ・ 𝑿𝒊: Criteria for describing the score: Factor (Feature) Long term forecast (longer than monthly) → portfolio construction Time information is omitted Problem Formulation: Learn the following model 𝑓
  • 12. Investment Horizon Minutes Hour Day Week Month Year Problem Formulation Time Series Cross Section Data Technical Fundamental Alternative Micro(Corporate)→Macro News、PoS、etc,… AI/Machine Learning Problem Formulation:Summary
  • 13. Contents 1. Introduction 2. Research Examples - Time series prediction - Cross sectional prediction 3. Conclusion
  • 14. Time Series × Technical × Supervised “Stock Price Prediction with Fluctuation Patterns Using Indexing Dynamic Time Warping and k∗-Nearest Neighbors”. Nakagawa et.al. (2018) New Frontiers in Artificial Intelligence https://doi.org/10.1007/978-3-319-93794-6_7
  • 15. Motivation Following nearest trend Following similar trend Momentum Patterns Predict the future price using the past price fluctuation patterns. Refine the “formation analysis” by machine learning. Nakagawa et.al. (2018) “Stock Price Prediction with Fluctuation Patterns Using Indexing Dynamic Time Warping and k∗-Nearest Neighbors”. New Frontiers in Artificial Intelligence https://doi.org/10.1007/978-3-319-93794-6_7
  • 16. Method - Outline Current pattern Umm... Similar!! 1. Measure 2. Predict ・How to measure the similarity? Indexing Dynamic Time Warping(IDTW) ・How to predict with extracted fluctuations? k*-Nearest Neighbors (k*-NN) We propose a method to predict future stock prices based on the shape of the past price fluctuations similar to the current.
  • 17. frequencymeasuring period 1. Decide the measuring period and frequency. 2. Consider the difference of price levels. Near? Distance 3. Define the distance between time-series. ✔ Daily closing prices within a month. ✔ Dynamic Time Warping (DTW) method How to measure the similarity between stock price movements? ✔ We index prices and adjust price levels. Method - IDTW
  • 18. ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Correspondence With Euclidean Distance 1 n n Sequences are aligned “one to one”. ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Correspondence With DTW 1 m n Nonlinear alignments are possible. Fixed Time Axis “Warped” Time Axis DTW is a method that warps the time axis to find the optimum correspondence and calculates the similarity between two-time series. What is DTW??
  • 19. Why do we use DTW?? ・ DTW can be applied to a pair of time-series of different lengths. ・ Similarity of DTW fits human intuition. ・ DTW is a standard method to measure the similarity among various time series data (UCR Datasets) What is DTW?? https://cran.r-project.org/web/packages/dtw/index.html R language package dtw
  • 20. Method - Outline Current pattern Umm... Similar!! 1. Measure 2. Predict ・How to measure the similarity? Indexing Dynamic Time Warping(IDTW) ・How to predict with extracted fluctuations? k*-Nearest Neighbors (k*-NN) We propose a method to predict future stock prices based on past price fluctuations similar to the current.
  • 21. Method - k-NN ✔ 𝑘-NN algorithm is one of the most fundamental non-parametric methods. Similarity 𝑘 = 1 𝑘 = 3 Prediction: +2.5% Prediction: (𝛼12.5 + 𝛼21.1 − 𝛼35.1)% Return on Next month (Labels) … +2.5% 𝑦1 +1.1% 𝑦2 −5.1% 𝑦3 −3.2% 𝑦4 0.15𝛽1 0.24𝛽2 0.41𝛽3 0.63𝛽4 𝛽𝑖 = 𝐼𝐷𝑇𝑊 𝑥0, 𝑥𝑖 Fluctuation pattern (Features) 𝑥0 𝑥1 𝑥2 𝑥3 𝑥4 ?% ✔ Prediction is the average of the values of 𝑘 nearest neighbors.
  • 22. Method - k*-NN ✔ 𝑘-NN algorithm is one of the most fundamental non-parametric methods. 𝑘 = 1 𝑘 = 3 Prediction: +2.5% Prediction: (𝛼12.5 + 𝛼21.1 − 𝛼35.1)% … +2.5% 𝑦1 +1.1% 𝑦2 −5.1% 𝑦3 −3.2% 𝑦4 ?% ✔ Prediction is the average of the values of 𝑘 nearest neighbors. 1. How many k should we use to predict? 2. How to decide the optimal weight alpha? 𝑘∗ -NN is a simple approach for calculating optimal 𝒌 and weight 𝜶. https://cran.r-project.org/web/packages/ksNN/index.html R language package ksNN
  • 23. Empirical Study ✔ To demonstrate the advantages of IDTW+k*NN, we analyze its performance using following major world stock indices. ✔ In-sample period : January 1989 ~ December 2005. Out-of-sample period : January 2006 ~ August 2017. ✔ Compared six methods combining DTW, DDTW, or IDTW with k-NN or k*-NN. Ticker Name Country Constituents TPX TOPIX Japan Companies listed on the First Section of the Tokyo Stock Exchange SPX S&P500 U.S. 500 large companies listed on the NYSE or NASDAQ CAC CAC40 France 40 companies listed on the Euronext Paris DAX DAX German 30 companies trading on the Frankfurt Stock Exchange UKX FTSE100 U.K. 100 companies listed on the London Stock Exchange
  • 24. Empirical Study Step 1. Calculate the similarity between month 𝑡 and all months from 1 to 𝑡 − 1 with DTW, DDTW, or IDTW. 𝑡 1 𝑡 − 1 Calculate each similarity t-3 t-2 t-1 t21 DTW DDTW or IDTW or
  • 25. Empirical Study Step 2. Predict the next month's return using k-NN or k*-NN. 6 120 t-2 27 11 Sorting by distance +1.5% 0.02 0.03 0.04 7.99 8.82 +1.9% -0.5% +3.0% -2.7% 𝛽 𝑦Next month’s returnsdistances k-NN k*-NN Predict or +𝟐. 𝟔%
  • 26. Empirical Study Step 3. If the predicted next month's return is positive, buy one unit of Index at the end of the month. If negative, sell one unit and calculate the return. Step 4. Proceed to the next month: 𝑡 = 𝑡 + 1. 𝑡 t t+1 𝑡 + 1 Index Price (Month End) 100 105 Predict +𝟐. 𝟔% +𝟓. 𝟎%Actual
  • 27. Results ✔ Cumulative returns for the six methods in TOPIX. The out-of-sample period is from January 2006 to August 2017. Red Line:Our method (IDTW+k*NN) Blue Line: Buy & Hold Index
  • 29. Results Table 1. The average accuracy of all years for each method. Table 2. The total return of each method.
  • 30. Results Table 3. Comparison with Momentum Strategy(Return) Table 4. Correlation with Momentum Strategy Here, Prediction in Step 2 is replaced by simple 1Mom: 1 month return and 12 Mom: 12 month return excluding the last month (12 Mom).
  • 31. Summary ✔ We proposed k*-NN with IDTW, a highly accurate stock prediction method. This method is inspired by formation analysis. ✔ An empirical analysis was conducted with major world indices and confirmed the following results: ・ IDTW is superior to DTW and DDTW ・ k*-NN is superior to k-NN. ・ IDTW-k*NN is the best prediction method.
  • 32. Time Series × Technical × Unsupervised “Stock price prediction using k‐medoids clustering with indexing dynamic time warping”. Nakagawa et.al. (2019) Electronics and Communications in Japan, 102(2), 3-8. https://doi.org/10.1002/ecj.12140
  • 33. Motivation Price movements that are likely to rise? ✔ Can we visualize good patterns for forecasting by clustering price fluctuation patterns? Price movements that are likely to fall? ✔ So how should we cluster the price fluctuation patterns? Nakagawa et.al. (2019). “Stock price prediction using k‐medoids clustering with indexing dynamic time warping”. Electronics and Communications in Japan, 102(2), 3-8.
  • 34. ・ Always looks for medoids in the data (Needs only similarity). ✔ Since the similarity was given by IDTW, we should decide the clustering method. k-means method k-medoids method How to Cluster Price Fluctuation? ・ Most well-used clustering method ・ Clustering data based on centroid ・ Easy to extent (x-means, k-means++) ・ Clustering data based on medoids (Data point closest to any data point in the cluster)
  • 35. 2007/1 2006/12 Clustering with k-Medoids 2004/5 Step1: Initialize medoids (𝑘: number of medoids) ✔ Randomly select 𝑘 = 3 medoids in the data
  • 36. 2007/1 2006/12 2004/5 IDTW Step2: Clustering ✔ Assign each data point to the cluster to which the closest medoid belong Step1: Initialize medoids (𝑘: number of medoids) Clustering with k-Medoids
  • 37. 2007/1 2006/12 Step3: Update medoids ✔ recalculate and update medoids in each cluster 2004/5 Step2: ClusteringStep1: Initialize medoids (𝑘: number of medoids) Repeat until the clusters do not change Clustering with k-Medoids
  • 38. Empirical Study ✔ Using the TOPIX, we cluster similar price fluctuation patterns and visualize good patterns for forecasting. ✔ In-sample period : January 1989 ~ December 2005. Out-of-sample period : January 2007 ~ March 2017. ✔ IDTW with k-medoids is compared to DTW with k-medoids ✔ The number of k are from 2 to 12. To decide k, we evaluate profitability of the following virtual trading.
  • 39. k-medoids is performed based on monthly daily price fluctuations up to the previous month. Identify the cluster for the current month. If there is more positive (negative) in the cluster, buy (sell). Step 2: Identify the cluster Step 3: Decide buy or sell Step1: Clustering Empirical Study
  • 40. ✔ From the viewpoint of profitability, the number of k is 𝟓 Table 5. Profitability and accuracy by number of clusters Results
  • 41. ✔ Visualization of 5 clusters (black line: Medoids, x axis: days, y axis: return) Reversal Momentum Momentum Results *The percentage in parentheses indicates the probability that stock price rises in the next month. *
  • 42. FlatReversal? Results ✔ Visualization of 5 clusters (black line: Medoids, x axis: days, y axis: return)
  • 43. ✔ Using k-Medoids + IDTW, we found that the number of clusters in the price fluctuation pattern was about 5 in TOPIX. ✔ Consider price fluctuation pattern clustering method (k-Medoids + IDTW). This is unsupervised approach. Summary ✔ Large fluctuations are tend to continue.
  • 44. Time Series × Technical × Supervised “Time-Series Gradient Boosting Tree for Stock Price Prediction”. Nakagawa and Yoshida (2020) International Journal of Data Mining, Modelling and Management (IJDMMM) To be appear Fundamental
  • 45. Motivation ✔ We improve a time series decision tree (TSDT) that uses only time series data. Nakagawa and Yoshida (2020) “Time Series Gradient Booting Tree for Stock Price Prediction”. International Journal of Data Mining, Modelling and Management (To be appear) (Time series) decision trees classify the (time series) data by sorting them down the tree and leaves provides the labels. Decision tree is easy to understand! Current
  • 46. Motivation We propose a time-series gradient boosting tree that uses a TSDT that adds features other than time-series. (1) Add usual features other than time series to TSDT (2) “Boost” this TSDT with gradient boosting technique
  • 47. Method – Time Series Gradient Boosting Tree ✔ Boosting is one of the methods called ensemble learning. ✔ Gradient boosting tree is most used in various data analysis contests (Kaggle) and has the best results. strong model weak model Ensemble learning is a method of building a strong (accurate) model by using multiple models called weak learners.
  • 48. Method – Time Series Gradient Boosting Tree Fit a new tree to the error Prettenhofer and Louppe(2014) Illustration of gradient boosting tree
  • 49. Empirical Study ✔ To demonstrate the advantages of time series gradient boosting tree, we analyze its performance using following major world stock indices. ✔ In-sample period : 10 years. Out-of-sample period : 10 years later until 2018/6 Ticker Name Country Data period TPX TOPIX Japan 1993/8~2018/6 SPX S&P500 U.S. 1990/1~2018/6 CAC CAC40 France 2001/6~2018/6 DAX DAX German 2001/7~2018/6 UKX FTSE100 U.K. 2000/1~2018/6
  • 50. Empirical Study ✔ Feature: Other Data :As of the end of month in each index, we use ✔ Model: Compared four methods combining Time Series Decision Tree (TSDT) and Time Series Gradient Boosting Tree (TSGBT) with and without index data. Data Description PER Net Profit/Market Value PBR Net Asset/Market Value MOM Stock returns over the past 12 months except for past month DIV Dividends paid in a year / Market Value ROE Net income/Net Assets Time Series Data:Daily price data within a month ✔ Teacher:1 month ahead return
  • 51. Results Table 6. The average accuracy of all years for each method. Table 7. The total return for each method. Improvement by adding features Improvement by boosting
  • 52. Summary ✔ We build a time series decision tree (TSDT) by introducing features other than time series data. ・Adding features other than time series data improves accuracy and profitability. ・Gradient boosting also improves accuracy and profitability ✔ Furthermore, we propose a time series gradient boosting tree (TSGBT) that “boosts” time-series decision tree. ✔ An empirical analysis was conducted with major world indices and confirmed the following results:
  • 53. 1. Introduction 2. Research Examples - Time series prediction - Cross sectional prediction 3. Conclusion Contents
  • 54. Cross Section × Technical × Supervised “Deep Factor Model”. Nakagawa et.al. (2018) 3rd Workshop on MIDAS@ECML-PKDD 2018 ([preprint] arXiv:1810.01278) “Cross-sectional Stock Price Prediction using Deep Learning for Actual Investment Management” Abe and Nakagawa (2020) AIBC2020 ([preprint] :arXiv:2002.06975) “Deep Learning for Multi-factor Models in Regional and Global Stock Markets” Abe and Nakagawa (2020) New Frontier in AI https://link.springer.com/chapter/10.1007/978-3-030-58790-1_6 Fundamental
  • 55. Cross Sectional Prediction ✔ Over 300 factors until 2012 [Harvey et al. 2017] ✔ In practice, we predict future relative stock returns (scores) by combining various factors ✔ Criteria for describing the score is called Factor in Finance For example, Market Value is one of the most famous factors, and this factor combined with simple sorting portfolio yields a positive return. Cross-sectional Investment Strategy ✔ finds stocks with relatively high scores based on certain criteria.
  • 56. ROE 1 Month Return ● ● ● Score Value Growth Quality Momentum ・Linear Relationship Factor Candidates Factor Classification by human Calculate Relative Score Cross-sectional Investment Strategy Both buying and selling are evaluated equally. Factors are evaluated monotonically.
  • 57. ROE 1 Month Return ● ● ● Score Value Growth Quality Momentum ・Linear Regression Factor Candidates Factor Classification by human Calculate Relative Score Cross-sectional Investment Strategy Both buying and selling are evaluated equally. Factors are evaluated monotonically. I would like to capture the non- linear relationship of factors from a large number of candidates by deep learning.
  • 58. Cross-sectional Investment Strategy w/ DL ROE 1 Month Return ● ● ● Factor Candidates ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ Extraction of effective factors in deep learning (without human judgment) Score ・Non-Linear Relationship
  • 59. 「Deep Factor Model」 Kei Nakagawa, Takumi Uchida and Tomohisa Aoshima, 3rd Workshop on MIDAS@ECML-PKDD 2018 ([preprint] arXiv:1810.01278) ✔ We check the profitability of cross-sectional investment strategy with deep learning ✔ We predict relative return using 80 factors and invest monthly. ✔ Investment Universe: TOPIX index constituents (from Apr 2006 to Mar 2016 monthly) Empirical Study No.1
  • 60. ✔ We use the 16 factors that are used relatively often in practice. Empirical Study No.1 Factor Candidates Formula Risk 60VOL Standard deviation of stock returns in the past 60 months BETA Regression coefficient of stock returns and market risk premium SKEW Skewness of stock returns in the past 60 months Quality ROE Net income/Net Assets ROA Operating Profit/Total Assets ACCRUALS Operating Cashflow ‐ Operating Profit LEVERAGE Total Liabilities / Total Assets Momentum 12-1MOM Stock returns in the past 12 months except for past month 1MOM Stock returns in the past month 60MOM Stock returns in the past 60 months Value PSR Sales/Market Value PER Net Income / Market Value PBR Net Assets / Market Value PCFR Operating Cashflow / Market Value Size CAP log(Market Value) ILLIQ average(Stock Returns/Trading Volume)
  • 61. ✔ Use 16 x 5 = 80 features for the last 5 points each quarter of 16 factors. Empirical Study No.1 ✔ We Learn the model using the training data for 120 sets. ✔ Training data (1 set) ・Feature: 80 features of 16 factors ・Teacher: One month ahead relative returns
  • 62. Model Description Deep Learning(Deep 1) We use full-connected network. The hidden layers are (80-50- 10),(80-80-50-50-10-10). We use the ReLU as the activation function and Adam for the optimization algorithm.Deep Learning(Deep 2) Linear Model (Linear Regression;LR) “sklearn.linear_model.LinearRegression” Support Vector Regression(SVR) “sklearn.svm.SVR” Random Forest(RF) “sklearn.ensemble.RandomForestRegressor” ✔ We use linear and nonlinear models as benchmarks. ✔ We make a long/short portfolio strategy for a net-zero investment to buy top stocks and to sell bottom stocks with equal weighting in quintile portfolios. Empirical Study No.1 ✔ For the quintile portfolio performance, we calculate the annualized average return, and Sharpe ratio.
  • 63. Results No.1 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% Deep 1 Deep 2 LR SVR RF Return✔ This graph shows the (annualized) returns of each methods The results of deep learning are better than other nonlinear models. The linear model is not bad either.
  • 64. Results No.1 ✔ This graph shows the Sharpe Ratio of each methods 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 Deep 1 Deep 2 LR SVR RF ShapeRatio The results of deep learning are better than other nonlinear models. The linear model is not bad either. Linear model is good if a poor nonlinear model is used.
  • 65. Empirical Study No.2 ✔ We check the profitability of cross-sectional investment strategy with deep learning ✔ We predict relative return using 33 factors and invest daily basis. ✔ Investment Universe: Cross-sectional Stock Price Prediction using Deep Learning for Actual Investment Management Masaya Abe and Kei Nakagawa, AIBC2020 ([preprint] arXiv: 2002.06975) TOPIX500 index constituents (from Apr 2013 to Mar 2017 daily)
  • 66. Empirical Study No.2 ✔ We use the 33 factors that are used relatively often in practice. No. Factor 1 Return from previous day 2 Return from 2 days ago 3 Return from 3 days ago 4 Return from 5 days ago 5 Return from 10 days ago 6 Return from 20 days ago 7 Return from 40 days ago 8 Return from 60 days ago 9 Average trading volume over the past 60 days 10 Average trading volume over the past 5 days/60 days 11 Average trading volume over the past 10 days/60 days 12 Average trading volume over the past 20 days/60 days 13 Change in operating income forecast from 5 days ago 14 Change in operating income forecast from 10 days ago 15 Change in operating income forecast from 20 days ago 16 Change in target stock price forecast from 5 days ago 17 Change in target stock price forecast from 10 days ago No. Factor 18 Change in target stock price forecast from 20 days ago 19 Book-value to Price Ratio 20 Earnings to Price Ratio 21 Dividend Yield 22 Sales to Price Ratio 23 Cashflow to Price Ratio 24 Return on Equity 25 Return on Asset 26 Return on Invested Capital 27 Accruals 28 Total Asset Turnover Rate 29 Current Ratio 30 Equity Ratio 31 Total Asset Growth Rate 32 Capital Expenditure Growth Rate 33 Investment to Asset - -
  • 67. ✔ We use linear and nonlinear models as benchmarks. Empirical Study No.2 Model Hidden Layer (Dropout) Epoch DNN_1 500 – 200 – 100 – 50 – 10 (50% – 40% – 30% – 20% – 10%) 20 DNN_2 500 – 200 – 100 – 50 – 10 (50% – 40% – 30% – 20% – 10%) 30 DNN_3 200 – 200 – 100 – 100 – 50 (50% – 50% – 30% –30% – 10%) 20 DNN_4 200 – 200 – 100 – 100 – 50 (50% – 50% – 30% – 30% – 10%) 30 DNN_5 300 – 300 – 150 – 150 – 50 (50% – 50% – 30% – 30% – 10%) 20 DNN_6 300 – 300 – 150 – 150 – 50 (50% – 50% – 30% – 30% – 10%) 30 ・ Deep Learning ・ Random Forest (RF) Model # of feature # of tree depth RF_1 11 1000 3 RF_2 11 1000 5 RF_3 11 1000 7 model regularization RR_1 0.1 RR_2 1.0 RR_3 10.0 ・ Linear Model (Ridge Regression;RR) ✔ We make a quintile long/short portfolio and calculate the annualized average return, and Sharpe ratio.
  • 68. For the epoch number, 20 is good. In this study, 30 is overfitting. Empirical Study No.2 ✔ This graph shows the (annualized) returns of each methods 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% Return For the level of return, the linear models(RR) are very good.
  • 69. Empirical Study No.2 ✔ This graph shows the Sharpe Ratio of each methods 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 ShapeRatio For the epoch number, 20 is good. In this study, 30 is overfitting. For the Shape ratio, DNNs are the very good.
  • 70. Empirical Study No.3 ✔ We check the profitability of cross-sectional investment strategy with deep learning ✔ We predict relative return using 20 factors and invest monthly. ✔ Investment Universe: Deep Learning for Multi-factor Models in Regional and Global Stock Markets Masaya Abe and Kei Nakagawa, New Frontiers in Artificial Intelligence (To be appear) World : MSCI WORLD North America: MSCI North Americas Europe: MSCI Europe & Middle East Asia Pacific: MSCI Pacific (from 2005 to 2017 monthly)
  • 71. Empirical Study No.3 ✔ We use the 20 factors that are used relatively often in practice. No. Factor No. Factor No. Factor 1 Book-to-market ratio 8 Return on invested capital 15 EPS Revision(1 month) 2 Earnings-to-price ratio 9 Accruals 16 EPS Revision(3 months) 3 Dividend yield 10 Total Asset Growth Rate 17 Past stock return(1 month) 4 Sales-to-price ratio 11 Current ratio 18 Past stock return(12 months) 5 Cash flow-to-price ratio 12 Equity ratio 19 Volatility 6 Return on equity 13 Total Asset Turnover Rate 20 Skewness 7 Return on asset 14 CAPEX Growth Rate
  • 72. ✔ We use linear and nonlinear models as benchmarks. Empirical Study No.3 Model Description Deep Learning We use full-connected network. The hidden layers are (150- 150-100-100-50-50). Dropout rate: (50% – 50% – 30% – 30% – 10% – 10%).We use the ReLU as the activation function and Adam for the optimization algorithm. Linear Model (Ridge Regression) “sklearn.linear_model.Ridge” Support Vector Regression “sklearn.svm.SVR” Random Forest “sklearn.ensemble.RandomForestRegressor” Gradient Boosting Tree “XGBRegressor” ✔ We make a quintile long/short portfolio and calculate the annualized average return, and Sharpe ratio.
  • 73. Results No.3 ✔ This graph shows the (annualized) returns of each methods 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 DNN RR XGB RF DNN RR XGB RF DNN RR XGB RF DNN RR XGB RF NA EU PF WLD Return ✓ DNN generally outperforms other methods in all regions. ✓ The performance decreases in the order of PF, EU, NA regardless of the prediction model.
  • 74. Results No.3 ✔ This graph shows the Sharpe Ratio of each methods 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 DNN RR XGB RF DNN RR XGB RF DNN RR XGB RF DNN RR XGB RF NA EU PF WLD ShapeRatio ✓ The highest Shape ratios are given by DNN in all regions. ✓ The performance decreases in the order of PF, EU, NA regardless of the prediction model.
  • 75. Summary DNN is generally superior to GB, RF, and RR in Japan and global markets → DNN shows promise as a skillful machine learning method in the cross section. The performance deteriorates in the order of Pacific, Europe, North America regardless of the prediction model. → The higher the market efficiency, the more difficult to earn returns
  • 76. Contents 1. Introduction 2. Research Examples - Time series prediction - Cross sectional prediction 3. Conclusion
  • 77. Problem Formulation • Time Series • Cross Section Data • Technical • Fundamental • Alternative AI/Machine Learning Conclusion If you have any question or comment, please feel free to Email me kei.nak.0315@gmail.com
  • 78. Disclaimer The Information in this presentation is provided for education and informational purposes only, without any express or implied warranty of any kind, including warranties of accuracy, completeness, or fitness for any particular purpose. The Information contained in or provided from or through this material is not intended to be and does not constitute financial advice, investment advice, trading advice or any other advice. The Information in this material is general in nature and is not specific to you the viewer or anyone else. If this document contains any information relating to the past performance of any securities or other investment strategies and products, you should note that past performance is not indicative of future results. Any market or investment view mentioned in this document is not intended to be investment research. Information contained in this document is subject to change without any obligation on our part to notify you of any change. If this document forms part of a presentation or is presented to you together with other documents and materials, this document should not be read in isolation and may not provide a full explanation of all of the topics presented and discussed. You should note that investment involves risk. The price of any securities or other investment product may go down as well as up. The information does not accept any liability or legal responsibility whatsoever for any decisions taken based upon the material.
  • 79. ∗ 𝑑(𝑥, 𝑦) = 𝑥 − 𝑦 ∗ 79 ▷ Indexing ▷ DTW Appendix Algorithm of IDTW
  • 81. 81 ▷ Step 1 ▷ Step 2 ▷ Step 3 Appendix Algorithm of k-medoids with IDTW
  • 82. 82 ▷ Time Series ▷ Data Appendix Algorithm of TSDT with usual data
  • 83. 83 Appendix Algorithm of TSGBT with usual data