SlideShare a Scribd company logo
Predictive Analytics for Alpha Generation and Risk Management
Yigal Jhirad
March 19, 2015
NVIDIA GTC 2015: Deep Learning and AI Conference
Silicon Valley
2
GTC 2015: Table of Contents
I. State-Space Models: State Instantiation
— Portfolio and Risk Management: Big Data/Real Time Sensitivity
— Applications - Cluster and Cointegration Analysis, Factor
Optimization
II. Parallel Implementation
— Coherence, Correlation, Cointegration, Clustering
III. Summary
IV. Author Biographies
DISCLAIMER: This presentation is for information purposes only. The presenter accepts no liability for the content of
this presentation, or for the consequences of any actions taken on the basis of the information provided. Although the
information in this presentation is considered to be accurate, this is not a representation that it is complete or should be
relied upon as a sole resource, as the information contained herein is subject to change.
3
GTC 2015: State-Space Models
 Portfolio/Risk Management
— Financials markets display discrete dynamic micro states and regimes
– Traditional valuation techniques may not be as effective, e.g. Options markets – Implied
Volatility vs. Jump Diffusion Regime Shifts
— Big Data : Time Series Tick Data, Interday and Intraday
— Predictive Analytics: Identifying Structural Breaks
— Alpha Generation, Risk Management, Market Impact
 Parallel Processing: APL/CPU and CUDA/GPU physics based modeling exploit hardware efficiently
— Array Based Processing paradigm Matrix/Vector thought process is key
— APL is a programming language whose quantum data object is an array, which is
fundamental to parallel processing and can leverage parallel processing across CPU’s
— CUDA leverages GPU Hardware
 Application in Econometrics and Applied Mathematics
— Monte Carlo Simulations — Fourier Analysis
— Principal Components — Optimization
— Cluster Analysis — Cointegration
4
GTC 2015: Cluster & Cointegration Analysis
 Cluster Analysis: A multivariate technique designed to identify relationships and cohesion
— Factor Analysis, Risk Model Development
 Correlation Analysis: Pairwise analysis of data across assets. Each pairwise comparison can be run in parallel.
— Use Correlation or Cointegration as primary input to cluster analysis
— Apply proprietary signal filter to remove selected data and reduce spurious correlations
5
Quantitative Cluster Analysis: Correlation
Application example: Correlation function removing N/A values
 Correlation measures the direction and strength of a linear relationship between variables. The Pearson
product moment correlation between two variables X and Y is calculated as:
 For N assets there are unique correlation pairs
 Given an N x M matrix A in which each row is a list of returns for a particular equity, return an N x N
matrix R in which each element is the scalar result of correlating each row of A to every other
— Each element of A may be an N/A value
— When processing a pair of rows, the calculation must include neither each N/A value nor the
corresponding element in the other row. This requires evaluating each pair separately.
— As a result the increased computational demand is more effectively implemented through a
parallel processing solution. As the matrix size increases the benefits of parallel processing become
more significant.
6
Cluster Analysis: Configuration
 Hardware Constraints
— Compute capability: 3.5
— Warp size: 32
— Number of shared memory banks: 32
— Coalescence capacity (4-byte words): 128 bytes = 32 contiguous words
— Max blocks / warps / threads per multiprocessor: 16 / 64 / 2048
— Max registers per multiprocessor: 64K
— Max shared memory per multiprocessor: 48K
 Software Constraints/Response
— Run 32 independent correlations per block for output coalescence
— Read 32 prices/returns per correlation for input coalescence
— Registers used per thread: 33 => max 15 blocks per SM @ 4 warps per block
— Shared memory used per block: 256 + 384 per warp => max 16 blocks per SM @ 4 warps per block
— Thread block configuration: 4 x 32 => 15 blocks / 60 warps / 1920 threads per SM
Tesla K20c
Each pair of rows produces a single correlation
Cluster Analysis: Kernel Operation
Input Output
Cluster Analysis: Kernel
Correlation computation
σ 𝑋− ഥ𝑋 𝑌−ത𝑌
σ 𝑋− ഥ𝑋 2
σ 𝑌−ത𝑌 2
Two passes through prices/returns in global memory required
First pass:
• Compute means of 𝑿 and 𝒀 ֜ ഥ𝑿, ഥ𝒀
Second pass:
• Subtract means from 𝑿 and 𝒀 ֜ 𝑿 ́, 𝒀 ́
• Sum 𝑿 ́ × 𝒀 ́, 𝑿 ́ × 𝑿 ́ and 𝒀 ́ × 𝒀 ́ ֜ σ𝑿 ́ 𝒀 ́, σ𝑿 ́2, σ𝒀 ́2
• Multiply resultant σ𝑿 ́ 𝒀 ́ by reciprocal square roots of σ𝑿 ́2 × σ𝒀 ́2
Cluster Analysis: Kernel First Pass
Computation of means
(row 33, col 0 of block grid)
 Repeat: read next section of
Prices/Returns into registers
 Ns necessary for NA
handling: “0” where
either Xs or Ys is NA;
“1” where Xs and Ys
are both valid
Cluster Analysis: Kernel First Pass
Computation of means
(row 33, col 0 of block grid)
 Repeat: sum values in
registers into shared
memory
+ + +
Cluster Analysis: Kernel First Pass
Computation of means
(row 33, col 0 of block grid)
 Sum rows to obtain totals for each row of Prices/Returns
 Save totals to shared memory
 Divide totals by N
Assets
÷ ÷
÷ ÷
Cluster Analysis: Kernel Second Pass
Computation of correlations
 Read section of Prices/
Returns into registers (reuse
registers from first pass)
 Ns necessary for NA
handling: “0” where
either Xs or Ys is NA;
“1” where Xs and Ys
are both valid
Cluster Analysis: Kernel Second Pass
Computation of correlations
 Subtract means of X and Y
− −
Cluster Analysis: Kernel Second Pass
Computation of correlations
 Multiply and sum (reuse
X, Y, N from first pass)
× ×
+ +
Cluster Analysis: Kernel Second Pass
Computation of correlations
 Multiply and sum (reuse
X, Y, N from first pass)
×
+
Cluster Analysis: Kernel Second Pass
Computation of correlations
 Sum rows to obtain σ of terms for each row of Prices/Returns
σ 𝑋 − ഥ𝑋 𝑌 − ത𝑌
σ 𝑋 − ഥ𝑋 2 σ 𝑌 − ത𝑌 2
Cluster Analysis: Kernel Second Pass
Computation of correlations
Assets
rsqrt(X2) × rsqrt(Y2) × XY
σ 𝑋 − ഥ𝑋 𝑌 − ത𝑌
σ 𝑋 − ഥ𝑋 2 σ 𝑌 − ത𝑌 2
Cluster Analysis: Kernel
Output to global memory : row 33, col 0 of block grid
Assets
Assets
Assets
Host/Global Memory
 Pinned-mapped memory eliminates transfer complexity, reduces overhead
 Kernel processing speed approximately 0.1 seconds for 586 assets × 120 periods
Cluster Analysis: Big Data, Real Time
Input Output
Overview
 Data source constantly streaming tick data
 Simple routine periodically aggregates ticks into VWAPs
or returns, producing an array of prices, one per ticker
 VWAPs/returns loaded into one column of input
Prices/Returns matrix in host memory at every interval
 Aggregation by period enables arbitrary real time speed,
limited by speed of GPU kernel algorithm
Cluster Analysis: Big Data, Real Time Streaming
data source
Routine to
periodically
aggregate
ticks into
VWAPs
UPG
To target
application
21
Summary
 Application of State –Space models using Intraday time series data
 Our approach is based on Time Series Cluster Analysis and Optimizations
— Identifying Structural Breaks, Patterns and Signals
— Coherency and Group membership
 Application in Portfolio Management for Alpha Generation and Risk Management
26
Author Biographies
 Yigal D. Jhirad, Senior Vice President, is Director of Quantitative and Derivatives
Strategies and a Portfolio Manager for Cohen & Steers’ options and real assets strategies.
Mr. Jhirad heads the firm’s Investment Risk Committee. He has 26 years of experience.
Prior to joining the firm in 2007, Mr. Jhirad was an executive director in the institutional
equities division of Morgan Stanley, where he headed the company’s portfolio and
derivatives strategies effort. He was responsible for developing, implementing and
marketing quantitative and derivatives products to a broad array of institutional clients,
including hedge funds, active and passive funds, pension funds and endowments. Mr.
Jhirad holds a BS from the Wharton School. He is a Financial Risk Manager (FRM), as
Certified by the Global Association of Risk Professionals.

More Related Content

What's hot

Machine Learning - Supervised Learning
Machine Learning - Supervised LearningMachine Learning - Supervised Learning
Machine Learning - Supervised Learning
Giorgio Alfredo Spedicato
 
Introduction to data structure and algorithms
Introduction to data structure and algorithmsIntroduction to data structure and algorithms
Introduction to data structure and algorithms
Research Scholar in Manonmaniam Sundaranar University
 
Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...
Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...
Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...
Jen Aman
 
Generalized Linear Models with H2O
Generalized Linear Models with H2O Generalized Linear Models with H2O
Generalized Linear Models with H2O
Sri Ambati
 
Machine Learning - Principles
Machine Learning - PrinciplesMachine Learning - Principles
Machine Learning - Principles
Giorgio Alfredo Spedicato
 
Kx for wine tasting
Kx for wine tastingKx for wine tasting
Kx for wine tasting
Mark Lefevre, CQF
 
Software Design Practices for Large-Scale Automation
Software Design Practices for Large-Scale AutomationSoftware Design Practices for Large-Scale Automation
Software Design Practices for Large-Scale Automation
Hao Xu
 
Quantitative finance in q
Quantitative finance in qQuantitative finance in q
Quantitative finance in q
Mark Lefevre, CQF
 
Generalized Linear Models in Spark MLlib and SparkR
Generalized Linear Models in Spark MLlib and SparkRGeneralized Linear Models in Spark MLlib and SparkR
Generalized Linear Models in Spark MLlib and SparkR
Databricks
 
Machine Learning in q/kdb+ - Teaching KDB to Read Japanese
Machine Learning in q/kdb+ - Teaching KDB to Read JapaneseMachine Learning in q/kdb+ - Teaching KDB to Read Japanese
Machine Learning in q/kdb+ - Teaching KDB to Read Japanese
Mark Lefevre, CQF
 
Circuit analysis i with matlab computing and simulink sim powersystems modeling
Circuit analysis i with matlab computing and simulink sim powersystems modelingCircuit analysis i with matlab computing and simulink sim powersystems modeling
Circuit analysis i with matlab computing and simulink sim powersystems modeling
Indra S Wahyudi
 
Towards an Incremental Schema-level Index for Distributed Linked Open Data G...
Towards an Incremental Schema-level Index  for Distributed Linked Open Data G...Towards an Incremental Schema-level Index  for Distributed Linked Open Data G...
Towards an Incremental Schema-level Index for Distributed Linked Open Data G...
Till Blume
 
Scalable Data Analysis in R Webinar Presentation
Scalable Data Analysis in R Webinar PresentationScalable Data Analysis in R Webinar Presentation
Scalable Data Analysis in R Webinar Presentation
Revolution Analytics
 
Chap2 slides
Chap2 slidesChap2 slides
Chap2 slides
BaliThorat1
 
Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...
Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...
Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...
Tiziano De Matteis
 
TINET_FRnOG_2008_public
TINET_FRnOG_2008_publicTINET_FRnOG_2008_public
TINET_FRnOG_2008_public
Davide Cherubini
 
Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...
Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...
Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...
IJERA Editor
 
Matlab day 1: Introduction to MATLAB
Matlab day 1: Introduction to MATLABMatlab day 1: Introduction to MATLAB
Matlab day 1: Introduction to MATLAB
reddyprasad reddyvari
 
Chap5 slides
Chap5 slidesChap5 slides
Chap5 slides
BaliThorat1
 
Block Sampling: Efficient Accurate Online Aggregation in MapReduce
Block Sampling: Efficient Accurate Online Aggregation in MapReduceBlock Sampling: Efficient Accurate Online Aggregation in MapReduce
Block Sampling: Efficient Accurate Online Aggregation in MapReduce
Vasia Kalavri
 

What's hot (20)

Machine Learning - Supervised Learning
Machine Learning - Supervised LearningMachine Learning - Supervised Learning
Machine Learning - Supervised Learning
 
Introduction to data structure and algorithms
Introduction to data structure and algorithmsIntroduction to data structure and algorithms
Introduction to data structure and algorithms
 
Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...
Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...
Massive Simulations In Spark: Distributed Monte Carlo For Global Health Forec...
 
Generalized Linear Models with H2O
Generalized Linear Models with H2O Generalized Linear Models with H2O
Generalized Linear Models with H2O
 
Machine Learning - Principles
Machine Learning - PrinciplesMachine Learning - Principles
Machine Learning - Principles
 
Kx for wine tasting
Kx for wine tastingKx for wine tasting
Kx for wine tasting
 
Software Design Practices for Large-Scale Automation
Software Design Practices for Large-Scale AutomationSoftware Design Practices for Large-Scale Automation
Software Design Practices for Large-Scale Automation
 
Quantitative finance in q
Quantitative finance in qQuantitative finance in q
Quantitative finance in q
 
Generalized Linear Models in Spark MLlib and SparkR
Generalized Linear Models in Spark MLlib and SparkRGeneralized Linear Models in Spark MLlib and SparkR
Generalized Linear Models in Spark MLlib and SparkR
 
Machine Learning in q/kdb+ - Teaching KDB to Read Japanese
Machine Learning in q/kdb+ - Teaching KDB to Read JapaneseMachine Learning in q/kdb+ - Teaching KDB to Read Japanese
Machine Learning in q/kdb+ - Teaching KDB to Read Japanese
 
Circuit analysis i with matlab computing and simulink sim powersystems modeling
Circuit analysis i with matlab computing and simulink sim powersystems modelingCircuit analysis i with matlab computing and simulink sim powersystems modeling
Circuit analysis i with matlab computing and simulink sim powersystems modeling
 
Towards an Incremental Schema-level Index for Distributed Linked Open Data G...
Towards an Incremental Schema-level Index  for Distributed Linked Open Data G...Towards an Incremental Schema-level Index  for Distributed Linked Open Data G...
Towards an Incremental Schema-level Index for Distributed Linked Open Data G...
 
Scalable Data Analysis in R Webinar Presentation
Scalable Data Analysis in R Webinar PresentationScalable Data Analysis in R Webinar Presentation
Scalable Data Analysis in R Webinar Presentation
 
Chap2 slides
Chap2 slidesChap2 slides
Chap2 slides
 
Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...
Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...
Keep Calm and React with Foresight: Strategies for Low-Latency and Energy-Eff...
 
TINET_FRnOG_2008_public
TINET_FRnOG_2008_publicTINET_FRnOG_2008_public
TINET_FRnOG_2008_public
 
Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...
Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...
Design of High speed Low Power Reversible Vedic multiplier and Reversible Div...
 
Matlab day 1: Introduction to MATLAB
Matlab day 1: Introduction to MATLABMatlab day 1: Introduction to MATLAB
Matlab day 1: Introduction to MATLAB
 
Chap5 slides
Chap5 slidesChap5 slides
Chap5 slides
 
Block Sampling: Efficient Accurate Online Aggregation in MapReduce
Block Sampling: Efficient Accurate Online Aggregation in MapReduceBlock Sampling: Efficient Accurate Online Aggregation in MapReduce
Block Sampling: Efficient Accurate Online Aggregation in MapReduce
 

Similar to Predictive Analytics for Alpha Generation and Risk Management

Workshop nwav 47 - LVS - Tool for Quantitative Data Analysis
Workshop nwav 47 - LVS - Tool for Quantitative Data AnalysisWorkshop nwav 47 - LVS - Tool for Quantitative Data Analysis
Workshop nwav 47 - LVS - Tool for Quantitative Data Analysis
Olga Scrivner
 
Performance Comparision of Machine Learning Algorithms
Performance Comparision of Machine Learning AlgorithmsPerformance Comparision of Machine Learning Algorithms
Performance Comparision of Machine Learning Algorithms
Dinusha Dilanka
 
AIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONAIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTION
IRJET Journal
 
Paper-Allstate-Claim-Severity
Paper-Allstate-Claim-SeverityPaper-Allstate-Claim-Severity
Paper-Allstate-Claim-Severity
Gon-soo Moon
 
Building and deploying analytics
Building and deploying analyticsBuilding and deploying analytics
Building and deploying analytics
Collin Bennett
 
DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictions
Anton Kulesh
 
Artificial intelligence and IoT
Artificial intelligence and IoTArtificial intelligence and IoT
Artificial intelligence and IoT
Veselin Pizurica
 
Machine_Learning_Co__
Machine_Learning_Co__Machine_Learning_Co__
Machine_Learning_Co__
Sitamarhi Institute of Technology
 
IRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms ComparisonIRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms Comparison
IRJET Journal
 
IRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms ComparisonIRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms Comparison
IRJET Journal
 
Machine Learning and AI: Core Methods and Applications
Machine Learning and AI: Core Methods and ApplicationsMachine Learning and AI: Core Methods and Applications
Machine Learning and AI: Core Methods and Applications
QuantUniversity
 
Big data 2.0, deep learning and financial Usecases
Big data 2.0, deep learning and financial UsecasesBig data 2.0, deep learning and financial Usecases
Big data 2.0, deep learning and financial Usecases
Arvind Rapaka
 
Reinforcement Learning to Mimic Portfolio Behavior
Reinforcement Learning to Mimic Portfolio BehaviorReinforcement Learning to Mimic Portfolio Behavior
Reinforcement Learning to Mimic Portfolio Behavior
Yigal D. Jhirad
 
CCC-Bicluster Analysis for Time Series Gene Expression Data
CCC-Bicluster Analysis for Time Series Gene Expression DataCCC-Bicluster Analysis for Time Series Gene Expression Data
CCC-Bicluster Analysis for Time Series Gene Expression Data
IRJET Journal
 
Experiments on Design Pattern Discovery
Experiments on Design Pattern DiscoveryExperiments on Design Pattern Discovery
Experiments on Design Pattern Discovery
Tim Menzies
 
Synthesis of analytical methods data driven decision-making
Synthesis of analytical methods data driven decision-makingSynthesis of analytical methods data driven decision-making
Synthesis of analytical methods data driven decision-making
Adam Doyle
 
Impact of Asymmetry of Internet Traffic for Heuristic Based Classification
Impact of Asymmetry of Internet Traffic for Heuristic Based ClassificationImpact of Asymmetry of Internet Traffic for Heuristic Based Classification
Impact of Asymmetry of Internet Traffic for Heuristic Based Classification
CSCJournals
 
ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016
Paolo Missier
 
Statistical Arbitrage
Statistical ArbitrageStatistical Arbitrage
Statistical Arbitrage
Shubham Patil
 
Data Mining Concepts - A survey paper
Data Mining Concepts - A survey paperData Mining Concepts - A survey paper
Data Mining Concepts - A survey paper
rahulmonikasharma
 

Similar to Predictive Analytics for Alpha Generation and Risk Management (20)

Workshop nwav 47 - LVS - Tool for Quantitative Data Analysis
Workshop nwav 47 - LVS - Tool for Quantitative Data AnalysisWorkshop nwav 47 - LVS - Tool for Quantitative Data Analysis
Workshop nwav 47 - LVS - Tool for Quantitative Data Analysis
 
Performance Comparision of Machine Learning Algorithms
Performance Comparision of Machine Learning AlgorithmsPerformance Comparision of Machine Learning Algorithms
Performance Comparision of Machine Learning Algorithms
 
AIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONAIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTION
 
Paper-Allstate-Claim-Severity
Paper-Allstate-Claim-SeverityPaper-Allstate-Claim-Severity
Paper-Allstate-Claim-Severity
 
Building and deploying analytics
Building and deploying analyticsBuilding and deploying analytics
Building and deploying analytics
 
DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictions
 
Artificial intelligence and IoT
Artificial intelligence and IoTArtificial intelligence and IoT
Artificial intelligence and IoT
 
Machine_Learning_Co__
Machine_Learning_Co__Machine_Learning_Co__
Machine_Learning_Co__
 
IRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms ComparisonIRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms Comparison
 
IRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms ComparisonIRJET- Supervised Learning Classification Algorithms Comparison
IRJET- Supervised Learning Classification Algorithms Comparison
 
Machine Learning and AI: Core Methods and Applications
Machine Learning and AI: Core Methods and ApplicationsMachine Learning and AI: Core Methods and Applications
Machine Learning and AI: Core Methods and Applications
 
Big data 2.0, deep learning and financial Usecases
Big data 2.0, deep learning and financial UsecasesBig data 2.0, deep learning and financial Usecases
Big data 2.0, deep learning and financial Usecases
 
Reinforcement Learning to Mimic Portfolio Behavior
Reinforcement Learning to Mimic Portfolio BehaviorReinforcement Learning to Mimic Portfolio Behavior
Reinforcement Learning to Mimic Portfolio Behavior
 
CCC-Bicluster Analysis for Time Series Gene Expression Data
CCC-Bicluster Analysis for Time Series Gene Expression DataCCC-Bicluster Analysis for Time Series Gene Expression Data
CCC-Bicluster Analysis for Time Series Gene Expression Data
 
Experiments on Design Pattern Discovery
Experiments on Design Pattern DiscoveryExperiments on Design Pattern Discovery
Experiments on Design Pattern Discovery
 
Synthesis of analytical methods data driven decision-making
Synthesis of analytical methods data driven decision-makingSynthesis of analytical methods data driven decision-making
Synthesis of analytical methods data driven decision-making
 
Impact of Asymmetry of Internet Traffic for Heuristic Based Classification
Impact of Asymmetry of Internet Traffic for Heuristic Based ClassificationImpact of Asymmetry of Internet Traffic for Heuristic Based Classification
Impact of Asymmetry of Internet Traffic for Heuristic Based Classification
 
ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016
 
Statistical Arbitrage
Statistical ArbitrageStatistical Arbitrage
Statistical Arbitrage
 
Data Mining Concepts - A survey paper
Data Mining Concepts - A survey paperData Mining Concepts - A survey paper
Data Mining Concepts - A survey paper
 

Recently uploaded

Using-Presentation-Software-to-the-Fullf.pptx
Using-Presentation-Software-to-the-Fullf.pptxUsing-Presentation-Software-to-the-Fullf.pptx
Using-Presentation-Software-to-the-Fullf.pptx
kainatfatyma9
 
ServiceNow CIS-ITSM Exam Dumps & Questions [2024]
ServiceNow CIS-ITSM Exam Dumps & Questions [2024]ServiceNow CIS-ITSM Exam Dumps & Questions [2024]
ServiceNow CIS-ITSM Exam Dumps & Questions [2024]
SkillCertProExams
 
Proposal: The Ark Project and The BEEP Inc
Proposal: The Ark Project and The BEEP IncProposal: The Ark Project and The BEEP Inc
Proposal: The Ark Project and The BEEP Inc
Raheem Muhammad
 
The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...
The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...
The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...
OECD Directorate for Financial and Enterprise Affairs
 
Why Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdf
Why Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdfWhy Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdf
Why Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdf
Ben Linders
 
怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样
怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样
怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样
kekzed
 
Prsentation for VIVA Welike project 1semester.pptx
Prsentation for VIVA Welike project 1semester.pptxPrsentation for VIVA Welike project 1semester.pptx
Prsentation for VIVA Welike project 1semester.pptx
prafulpawar29
 
Legislation And Regulations For Import, Manufacture,.pptx
Legislation And Regulations For Import, Manufacture,.pptxLegislation And Regulations For Import, Manufacture,.pptx
Legislation And Regulations For Import, Manufacture,.pptx
Charmi13
 
The Intersection between Competition and Data Privacy – COLANGELO – June 2024...
The Intersection between Competition and Data Privacy – COLANGELO – June 2024...The Intersection between Competition and Data Privacy – COLANGELO – June 2024...
The Intersection between Competition and Data Privacy – COLANGELO – June 2024...
OECD Directorate for Financial and Enterprise Affairs
 
一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理
一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理
一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理
gfysze
 
IEEE CIS Webinar Sustainable futures.pdf
IEEE CIS Webinar Sustainable futures.pdfIEEE CIS Webinar Sustainable futures.pdf
IEEE CIS Webinar Sustainable futures.pdf
Claudio Gallicchio
 
2 December UAE National Day - United Arab Emirates
2 December UAE National Day - United Arab Emirates2 December UAE National Day - United Arab Emirates
2 December UAE National Day - United Arab Emirates
UAE Ppt
 
The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...
The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...
The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...
OECD Directorate for Financial and Enterprise Affairs
 
Disaster Management project for holidays homework and other uses
Disaster Management project for holidays homework and other usesDisaster Management project for holidays homework and other uses
Disaster Management project for holidays homework and other uses
RIDHIMAGARG21
 
The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...
The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...
The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...
OECD Directorate for Financial and Enterprise Affairs
 
ACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPE
ACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPEACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPE
ACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPE
Charmi13
 
BRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdf
BRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdfBRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdf
BRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdf
Robin Haunschild
 
Gamify it until you make it Improving Agile Development and Operations with ...
Gamify it until you make it  Improving Agile Development and Operations with ...Gamify it until you make it  Improving Agile Development and Operations with ...
Gamify it until you make it Improving Agile Development and Operations with ...
Ben Linders
 
Genesis chapter 3 Isaiah Scudder.pptx
Genesis    chapter 3 Isaiah Scudder.pptxGenesis    chapter 3 Isaiah Scudder.pptx
Genesis chapter 3 Isaiah Scudder.pptx
FamilyWorshipCenterD
 

Recently uploaded (19)

Using-Presentation-Software-to-the-Fullf.pptx
Using-Presentation-Software-to-the-Fullf.pptxUsing-Presentation-Software-to-the-Fullf.pptx
Using-Presentation-Software-to-the-Fullf.pptx
 
ServiceNow CIS-ITSM Exam Dumps & Questions [2024]
ServiceNow CIS-ITSM Exam Dumps & Questions [2024]ServiceNow CIS-ITSM Exam Dumps & Questions [2024]
ServiceNow CIS-ITSM Exam Dumps & Questions [2024]
 
Proposal: The Ark Project and The BEEP Inc
Proposal: The Ark Project and The BEEP IncProposal: The Ark Project and The BEEP Inc
Proposal: The Ark Project and The BEEP Inc
 
The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...
The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...
The Intersection between Competition and Data Privacy – OECD – June 2024 OECD...
 
Why Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdf
Why Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdfWhy Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdf
Why Psychological Safety Matters for Software Teams - ACE 2024 - Ben Linders.pdf
 
怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样
怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样
怎么办理(lincoln学位证书)英国林肯大学毕业证文凭学位证书原版一模一样
 
Prsentation for VIVA Welike project 1semester.pptx
Prsentation for VIVA Welike project 1semester.pptxPrsentation for VIVA Welike project 1semester.pptx
Prsentation for VIVA Welike project 1semester.pptx
 
Legislation And Regulations For Import, Manufacture,.pptx
Legislation And Regulations For Import, Manufacture,.pptxLegislation And Regulations For Import, Manufacture,.pptx
Legislation And Regulations For Import, Manufacture,.pptx
 
The Intersection between Competition and Data Privacy – COLANGELO – June 2024...
The Intersection between Competition and Data Privacy – COLANGELO – June 2024...The Intersection between Competition and Data Privacy – COLANGELO – June 2024...
The Intersection between Competition and Data Privacy – COLANGELO – June 2024...
 
一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理
一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理
一比一原版(unc毕业证书)美国北卡罗来纳大学教堂山分校毕业证如何办理
 
IEEE CIS Webinar Sustainable futures.pdf
IEEE CIS Webinar Sustainable futures.pdfIEEE CIS Webinar Sustainable futures.pdf
IEEE CIS Webinar Sustainable futures.pdf
 
2 December UAE National Day - United Arab Emirates
2 December UAE National Day - United Arab Emirates2 December UAE National Day - United Arab Emirates
2 December UAE National Day - United Arab Emirates
 
The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...
The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...
The Intersection between Competition and Data Privacy – CAPEL – June 2024 OEC...
 
Disaster Management project for holidays homework and other uses
Disaster Management project for holidays homework and other usesDisaster Management project for holidays homework and other uses
Disaster Management project for holidays homework and other uses
 
The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...
The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...
The Intersection between Competition and Data Privacy – KEMP – June 2024 OECD...
 
ACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPE
ACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPEACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPE
ACTIVE IMPLANTABLE MEDICAL DEVICE IN EUROPE
 
BRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdf
BRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdfBRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdf
BRIC_2024_2024-06-06-11:30-haunschild_archival_version.pdf
 
Gamify it until you make it Improving Agile Development and Operations with ...
Gamify it until you make it  Improving Agile Development and Operations with ...Gamify it until you make it  Improving Agile Development and Operations with ...
Gamify it until you make it Improving Agile Development and Operations with ...
 
Genesis chapter 3 Isaiah Scudder.pptx
Genesis    chapter 3 Isaiah Scudder.pptxGenesis    chapter 3 Isaiah Scudder.pptx
Genesis chapter 3 Isaiah Scudder.pptx
 

Predictive Analytics for Alpha Generation and Risk Management

  • 1. Predictive Analytics for Alpha Generation and Risk Management Yigal Jhirad March 19, 2015 NVIDIA GTC 2015: Deep Learning and AI Conference Silicon Valley
  • 2. 2 GTC 2015: Table of Contents I. State-Space Models: State Instantiation — Portfolio and Risk Management: Big Data/Real Time Sensitivity — Applications - Cluster and Cointegration Analysis, Factor Optimization II. Parallel Implementation — Coherence, Correlation, Cointegration, Clustering III. Summary IV. Author Biographies DISCLAIMER: This presentation is for information purposes only. The presenter accepts no liability for the content of this presentation, or for the consequences of any actions taken on the basis of the information provided. Although the information in this presentation is considered to be accurate, this is not a representation that it is complete or should be relied upon as a sole resource, as the information contained herein is subject to change.
  • 3. 3 GTC 2015: State-Space Models  Portfolio/Risk Management — Financials markets display discrete dynamic micro states and regimes – Traditional valuation techniques may not be as effective, e.g. Options markets – Implied Volatility vs. Jump Diffusion Regime Shifts — Big Data : Time Series Tick Data, Interday and Intraday — Predictive Analytics: Identifying Structural Breaks — Alpha Generation, Risk Management, Market Impact  Parallel Processing: APL/CPU and CUDA/GPU physics based modeling exploit hardware efficiently — Array Based Processing paradigm Matrix/Vector thought process is key — APL is a programming language whose quantum data object is an array, which is fundamental to parallel processing and can leverage parallel processing across CPU’s — CUDA leverages GPU Hardware  Application in Econometrics and Applied Mathematics — Monte Carlo Simulations — Fourier Analysis — Principal Components — Optimization — Cluster Analysis — Cointegration
  • 4. 4 GTC 2015: Cluster & Cointegration Analysis  Cluster Analysis: A multivariate technique designed to identify relationships and cohesion — Factor Analysis, Risk Model Development  Correlation Analysis: Pairwise analysis of data across assets. Each pairwise comparison can be run in parallel. — Use Correlation or Cointegration as primary input to cluster analysis — Apply proprietary signal filter to remove selected data and reduce spurious correlations
  • 5. 5 Quantitative Cluster Analysis: Correlation Application example: Correlation function removing N/A values  Correlation measures the direction and strength of a linear relationship between variables. The Pearson product moment correlation between two variables X and Y is calculated as:  For N assets there are unique correlation pairs  Given an N x M matrix A in which each row is a list of returns for a particular equity, return an N x N matrix R in which each element is the scalar result of correlating each row of A to every other — Each element of A may be an N/A value — When processing a pair of rows, the calculation must include neither each N/A value nor the corresponding element in the other row. This requires evaluating each pair separately. — As a result the increased computational demand is more effectively implemented through a parallel processing solution. As the matrix size increases the benefits of parallel processing become more significant.
  • 6. 6 Cluster Analysis: Configuration  Hardware Constraints — Compute capability: 3.5 — Warp size: 32 — Number of shared memory banks: 32 — Coalescence capacity (4-byte words): 128 bytes = 32 contiguous words — Max blocks / warps / threads per multiprocessor: 16 / 64 / 2048 — Max registers per multiprocessor: 64K — Max shared memory per multiprocessor: 48K  Software Constraints/Response — Run 32 independent correlations per block for output coalescence — Read 32 prices/returns per correlation for input coalescence — Registers used per thread: 33 => max 15 blocks per SM @ 4 warps per block — Shared memory used per block: 256 + 384 per warp => max 16 blocks per SM @ 4 warps per block — Thread block configuration: 4 x 32 => 15 blocks / 60 warps / 1920 threads per SM Tesla K20c
  • 7. Each pair of rows produces a single correlation Cluster Analysis: Kernel Operation Input Output
  • 8. Cluster Analysis: Kernel Correlation computation σ 𝑋− ഥ𝑋 𝑌−ത𝑌 σ 𝑋− ഥ𝑋 2 σ 𝑌−ത𝑌 2 Two passes through prices/returns in global memory required First pass: • Compute means of 𝑿 and 𝒀 ֜ ഥ𝑿, ഥ𝒀 Second pass: • Subtract means from 𝑿 and 𝒀 ֜ 𝑿 ́, 𝒀 ́ • Sum 𝑿 ́ × 𝒀 ́, 𝑿 ́ × 𝑿 ́ and 𝒀 ́ × 𝒀 ́ ֜ σ𝑿 ́ 𝒀 ́, σ𝑿 ́2, σ𝒀 ́2 • Multiply resultant σ𝑿 ́ 𝒀 ́ by reciprocal square roots of σ𝑿 ́2 × σ𝒀 ́2
  • 9. Cluster Analysis: Kernel First Pass Computation of means (row 33, col 0 of block grid)  Repeat: read next section of Prices/Returns into registers  Ns necessary for NA handling: “0” where either Xs or Ys is NA; “1” where Xs and Ys are both valid
  • 10. Cluster Analysis: Kernel First Pass Computation of means (row 33, col 0 of block grid)  Repeat: sum values in registers into shared memory + + +
  • 11. Cluster Analysis: Kernel First Pass Computation of means (row 33, col 0 of block grid)  Sum rows to obtain totals for each row of Prices/Returns  Save totals to shared memory  Divide totals by N Assets ÷ ÷ ÷ ÷
  • 12. Cluster Analysis: Kernel Second Pass Computation of correlations  Read section of Prices/ Returns into registers (reuse registers from first pass)  Ns necessary for NA handling: “0” where either Xs or Ys is NA; “1” where Xs and Ys are both valid
  • 13. Cluster Analysis: Kernel Second Pass Computation of correlations  Subtract means of X and Y − −
  • 14. Cluster Analysis: Kernel Second Pass Computation of correlations  Multiply and sum (reuse X, Y, N from first pass) × × + +
  • 15. Cluster Analysis: Kernel Second Pass Computation of correlations  Multiply and sum (reuse X, Y, N from first pass) × +
  • 16. Cluster Analysis: Kernel Second Pass Computation of correlations  Sum rows to obtain σ of terms for each row of Prices/Returns σ 𝑋 − ഥ𝑋 𝑌 − ത𝑌 σ 𝑋 − ഥ𝑋 2 σ 𝑌 − ത𝑌 2
  • 17. Cluster Analysis: Kernel Second Pass Computation of correlations Assets rsqrt(X2) × rsqrt(Y2) × XY σ 𝑋 − ഥ𝑋 𝑌 − ത𝑌 σ 𝑋 − ഥ𝑋 2 σ 𝑌 − ത𝑌 2
  • 18. Cluster Analysis: Kernel Output to global memory : row 33, col 0 of block grid Assets Assets Assets
  • 19. Host/Global Memory  Pinned-mapped memory eliminates transfer complexity, reduces overhead  Kernel processing speed approximately 0.1 seconds for 586 assets × 120 periods Cluster Analysis: Big Data, Real Time Input Output
  • 20. Overview  Data source constantly streaming tick data  Simple routine periodically aggregates ticks into VWAPs or returns, producing an array of prices, one per ticker  VWAPs/returns loaded into one column of input Prices/Returns matrix in host memory at every interval  Aggregation by period enables arbitrary real time speed, limited by speed of GPU kernel algorithm Cluster Analysis: Big Data, Real Time Streaming data source Routine to periodically aggregate ticks into VWAPs UPG To target application
  • 21. 21 Summary  Application of State –Space models using Intraday time series data  Our approach is based on Time Series Cluster Analysis and Optimizations — Identifying Structural Breaks, Patterns and Signals — Coherency and Group membership  Application in Portfolio Management for Alpha Generation and Risk Management
  • 22.
  • 23.
  • 24.
  • 25.
  • 26. 26 Author Biographies  Yigal D. Jhirad, Senior Vice President, is Director of Quantitative and Derivatives Strategies and a Portfolio Manager for Cohen & Steers’ options and real assets strategies. Mr. Jhirad heads the firm’s Investment Risk Committee. He has 26 years of experience. Prior to joining the firm in 2007, Mr. Jhirad was an executive director in the institutional equities division of Morgan Stanley, where he headed the company’s portfolio and derivatives strategies effort. He was responsible for developing, implementing and marketing quantitative and derivatives products to a broad array of institutional clients, including hedge funds, active and passive funds, pension funds and endowments. Mr. Jhirad holds a BS from the Wharton School. He is a Financial Risk Manager (FRM), as Certified by the Global Association of Risk Professionals.