SlideShare a Scribd company logo
1 of 10
A Machine Learning (ML) Crystal
Ball? How We Predict Future Outcomes
Using a Temporal Fusion Transformer
Model
Created by: Salvatore Tirabassi
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
Our data science and analytics teams handle and apply lots of data for
insightful decision-making. Last year, I presented the data science team with
a challenge: use historical data to predict a key business driver for each of
the next 8 periods.
We wanted to have a data-driven preview of what we might see in the in each
of the next eight periods so that we could anticipate the actual outcome and
make better decisions with an eight-period window.
Document Copyright 2023
The data science team went to work researching ways we could do this and
tested a few different methodologies. We had lots of input data from our own
and public sources to feed any model we wanted to test, which worked well for
us. With that said, we had low expectations about finding a predictive model
that produced anything reliable.
Testing different algorithms is always our approach. For the semi-technical
readers, before settling on Temporal Fusion Transformer (TFT), the algorithms
we tested included ARIMA, VAR, GARCH, ARCH models (univariate), Prophet, NHits,
and Nbeats. TFT is an attention-based deep learning neural network algorithm.
Using a mix of inputs, it produces a forecast over multiple periods in a future
time horizon that you can determine. You can predict days, weeks, months
quarters (really any interval is possible) into the future. Your choice.
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
The picture below shows the concept of how TFT works.
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
Source: Bryan Lim,
Sercan Ö. Arık, Nicolas
Loeff, Tomas Pfister.
“Temporal Fusion
Transformers for
interpretable multi-
horizon time series
forecasting.”
A continuous improvement process best describes how we developed and
continue to refine the model. It’s a never ending process of
improvement, as a true crystal ball is never achieved.
These are the four stages of development we went through after
choosing TFT as our algorithm:
Stage 1a: Selecting all logical observed inputs and test how they drive
the model. We started with over 100 and the final model only used 20.
Go to Stage 1b as needed.
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
Stage 1b: Refining the time intervals of the observed inputs. Since the
inputs might come in varying time intervals (daily, weekly, monthly and
quarterly, etc…), we needed to find methods to standardize them. You should
choose an interval that matches the decision-making forecast you are
producing, if you can. Go back to Stage 1a as needed.
Stage 2: Model iteration and improvement. Complete back testing. Examine early
predictions. Go back to Stages 1a and 1b as needed. At this point, you have
probably settled on one or two of the most promising algorithms.
Stage 3: Begin using in production and comparing predictions to the future
periods as they unfold. Learn and refine by going back to any previous stage as
needed.
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
Stage 4: Continuous improvement loop. Write a long-term road map. Test new
inputs as they are presented. Continuous scrutiny of the predictions against
what actually happens – learn and make changes by going back to any previous
stage as needed.
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
Note that at any of the stages of development, you can use a TFT encoder decoder to measure the importance of
different inputs in the algorithm to learn which ones have the most impact on your prediction.
Below are the results of our model. The orange line is the actual result of
the key driver and the blue line is the prediction of the key driver that was
made 8 periods prior. The area to the right without the orange line is the
next 8-period forecast.
So, at Period 19, we use the blue line forecast to take action based on what
Periods 20-27 tell us. When we reach Period 20, we evaluate the updated
forecast and we make a decisions accordingly for the future periods. This way,
we have a rolling 8-period prediction/decision cycle.
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
Document Copyright 2023
A Machine Learning (ML) Crystal
Ball? How We Predict Future
Outcomes Using a Temporal Fusion
Transformer Model
As you can see the model has
been refined to a level that
it makes useful predictions
and handles volatility of the
prediction with some
reliability. Right now, we
don’t use this to predict the
future down to the exact
number, which would be ideal,
but we use it for a
directional understanding of
where things are headed so we
can make better decisions at
Visit my content
https://salvatoretirabassi.substack.com
https://linkedin/in/stirabassi
https://tirabassi.com/
https://twitter.com/saltirabassi
https://www.instagram.com/salvatoretirabassi/
https://www.facebook.com/SalvatoreTirabassi
Document Copyright 2023

More Related Content

Similar to A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model

Time Series Weather Forecasting Techniques: Literature Survey
Time Series Weather Forecasting Techniques: Literature SurveyTime Series Weather Forecasting Techniques: Literature Survey
Time Series Weather Forecasting Techniques: Literature SurveyIRJET Journal
 
No estimates - 10 new principles for testing
No estimates  - 10 new principles for testingNo estimates  - 10 new principles for testing
No estimates - 10 new principles for testingVasco Duarte
 
4_5821003798000177081.pdf
4_5821003798000177081.pdf4_5821003798000177081.pdf
4_5821003798000177081.pdfozgenn4242
 
Six Sigma | Arrelic Insights
Six Sigma | Arrelic InsightsSix Sigma | Arrelic Insights
Six Sigma | Arrelic InsightsArrelic
 
Using Simulation and Optimization to Inform Hiring Decisions_Ginger_Castle
Using Simulation and Optimization to Inform Hiring Decisions_Ginger_CastleUsing Simulation and Optimization to Inform Hiring Decisions_Ginger_Castle
Using Simulation and Optimization to Inform Hiring Decisions_Ginger_CastleGinger Castle
 
AlgoB – Cryptocurrency price prediction system using LSTM
AlgoB – Cryptocurrency price prediction system using LSTMAlgoB – Cryptocurrency price prediction system using LSTM
AlgoB – Cryptocurrency price prediction system using LSTMIRJET Journal
 
Machine Learning for Forecasting: From Data to Deployment
Machine Learning for Forecasting: From Data to DeploymentMachine Learning for Forecasting: From Data to Deployment
Machine Learning for Forecasting: From Data to DeploymentAnant Agarwal
 
Better Living Through Analytics - Strategies for Data Decisions
Better Living Through Analytics - Strategies for Data DecisionsBetter Living Through Analytics - Strategies for Data Decisions
Better Living Through Analytics - Strategies for Data DecisionsProduct School
 
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 StocksLawrence Takeuchi
 
Simulation theory
Simulation theorySimulation theory
Simulation theoryAbu Bashar
 
Descriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docx
Descriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docxDescriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docx
Descriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docxcuddietheresa
 
Modeling and simulation
Modeling and simulationModeling and simulation
Modeling and simulationPayel Rani
 
Deep Learning Applications in Finance.pdf
Deep Learning Applications in Finance.pdfDeep Learning Applications in Finance.pdf
Deep Learning Applications in Finance.pdfVanessa Bridge
 
Machine learning at b.e.s.t. summer university
Machine learning  at b.e.s.t. summer universityMachine learning  at b.e.s.t. summer university
Machine learning at b.e.s.t. summer universityLászló Kovács
 
Turnover Prediction of Shares Using Data Mining Techniques : A Case Study
Turnover Prediction of Shares Using Data Mining Techniques : A Case Study Turnover Prediction of Shares Using Data Mining Techniques : A Case Study
Turnover Prediction of Shares Using Data Mining Techniques : A Case Study csandit
 
Aditya report finaL
Aditya report finaLAditya report finaL
Aditya report finaL2767882
 
Big Data & Machine Learning - TDC2013 Sao Paulo
Big Data & Machine Learning - TDC2013 Sao PauloBig Data & Machine Learning - TDC2013 Sao Paulo
Big Data & Machine Learning - TDC2013 Sao PauloOCTO Technology
 

Similar to A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model (20)

Time Series Weather Forecasting Techniques: Literature Survey
Time Series Weather Forecasting Techniques: Literature SurveyTime Series Weather Forecasting Techniques: Literature Survey
Time Series Weather Forecasting Techniques: Literature Survey
 
No estimates - 10 new principles for testing
No estimates  - 10 new principles for testingNo estimates  - 10 new principles for testing
No estimates - 10 new principles for testing
 
4_5821003798000177081.pdf
4_5821003798000177081.pdf4_5821003798000177081.pdf
4_5821003798000177081.pdf
 
Six Sigma | Arrelic Insights
Six Sigma | Arrelic InsightsSix Sigma | Arrelic Insights
Six Sigma | Arrelic Insights
 
Using Simulation and Optimization to Inform Hiring Decisions_Ginger_Castle
Using Simulation and Optimization to Inform Hiring Decisions_Ginger_CastleUsing Simulation and Optimization to Inform Hiring Decisions_Ginger_Castle
Using Simulation and Optimization to Inform Hiring Decisions_Ginger_Castle
 
AlgoB – Cryptocurrency price prediction system using LSTM
AlgoB – Cryptocurrency price prediction system using LSTMAlgoB – Cryptocurrency price prediction system using LSTM
AlgoB – Cryptocurrency price prediction system using LSTM
 
Machine Learning for Forecasting: From Data to Deployment
Machine Learning for Forecasting: From Data to DeploymentMachine Learning for Forecasting: From Data to Deployment
Machine Learning for Forecasting: From Data to Deployment
 
Better Living Through Analytics - Strategies for Data Decisions
Better Living Through Analytics - Strategies for Data DecisionsBetter Living Through Analytics - Strategies for Data Decisions
Better Living Through Analytics - Strategies for Data Decisions
 
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
 
Note about CFAR-m
Note about CFAR-mNote about CFAR-m
Note about CFAR-m
 
Note about CFAR-m
Note about CFAR-mNote about CFAR-m
Note about CFAR-m
 
Simulation theory
Simulation theorySimulation theory
Simulation theory
 
Descriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docx
Descriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docxDescriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docx
Descriptionsordernametypeformatvallabvarlab1location_idint8.0gNum.docx
 
Modeling and simulation
Modeling and simulationModeling and simulation
Modeling and simulation
 
Test for AI model
Test for AI modelTest for AI model
Test for AI model
 
Deep Learning Applications in Finance.pdf
Deep Learning Applications in Finance.pdfDeep Learning Applications in Finance.pdf
Deep Learning Applications in Finance.pdf
 
Machine learning at b.e.s.t. summer university
Machine learning  at b.e.s.t. summer universityMachine learning  at b.e.s.t. summer university
Machine learning at b.e.s.t. summer university
 
Turnover Prediction of Shares Using Data Mining Techniques : A Case Study
Turnover Prediction of Shares Using Data Mining Techniques : A Case Study Turnover Prediction of Shares Using Data Mining Techniques : A Case Study
Turnover Prediction of Shares Using Data Mining Techniques : A Case Study
 
Aditya report finaL
Aditya report finaLAditya report finaL
Aditya report finaL
 
Big Data & Machine Learning - TDC2013 Sao Paulo
Big Data & Machine Learning - TDC2013 Sao PauloBig Data & Machine Learning - TDC2013 Sao Paulo
Big Data & Machine Learning - TDC2013 Sao Paulo
 

More from CFO Pro+Analytics

Managing a Loan Portfolio with Great Analytical Tools.pdf
Managing a Loan Portfolio with Great Analytical Tools.pdfManaging a Loan Portfolio with Great Analytical Tools.pdf
Managing a Loan Portfolio with Great Analytical Tools.pdfCFO Pro+Analytics
 
Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...
Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...
Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...CFO Pro+Analytics
 
What's Really Going on With Consumer Balance Sheets?
What's Really Going on With Consumer Balance Sheets?What's Really Going on With Consumer Balance Sheets?
What's Really Going on With Consumer Balance Sheets?CFO Pro+Analytics
 
Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...
Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...
Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...CFO Pro+Analytics
 
Applying Predictive Modeling to Crypto Futures Trading
Applying Predictive Modeling to Crypto Futures TradingApplying Predictive Modeling to Crypto Futures Trading
Applying Predictive Modeling to Crypto Futures TradingCFO Pro+Analytics
 
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...CFO Pro+Analytics
 
Cracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptx
Cracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptxCracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptx
Cracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptxCFO Pro+Analytics
 
Consumer Credit Card Interest Savings in a Decreasing Rate Environment.pptx
Consumer Credit Card Interest Savings in a Decreasing Rate Environment.pptxConsumer Credit Card Interest Savings in a Decreasing Rate Environment.pptx
Consumer Credit Card Interest Savings in a Decreasing Rate Environment.pptxCFO Pro+Analytics
 
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...CFO Pro+Analytics
 
Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...
Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...
Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...CFO Pro+Analytics
 
Decoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptx
Decoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptxDecoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptx
Decoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptxCFO Pro+Analytics
 
How we replaced an implementation of workday adaptive planning enterprise man...
How we replaced an implementation of workday adaptive planning enterprise man...How we replaced an implementation of workday adaptive planning enterprise man...
How we replaced an implementation of workday adaptive planning enterprise man...CFO Pro+Analytics
 
LCV-Understanding lifetime customer value-.pptx
LCV-Understanding lifetime customer value-.pptxLCV-Understanding lifetime customer value-.pptx
LCV-Understanding lifetime customer value-.pptxCFO Pro+Analytics
 
Unlocking consumer loan pricing a deep dive into survival regression algorith...
Unlocking consumer loan pricing a deep dive into survival regression algorith...Unlocking consumer loan pricing a deep dive into survival regression algorith...
Unlocking consumer loan pricing a deep dive into survival regression algorith...CFO Pro+Analytics
 
Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...
Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...
Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...CFO Pro+Analytics
 
Using a Survival Regression to Evaluate.pptx
Using a Survival Regression to Evaluate.pptxUsing a Survival Regression to Evaluate.pptx
Using a Survival Regression to Evaluate.pptxCFO Pro+Analytics
 
Unraveling the Complexities of Consumer Balances Sheets.pptx
Unraveling the Complexities of Consumer Balances Sheets.pptxUnraveling the Complexities of Consumer Balances Sheets.pptx
Unraveling the Complexities of Consumer Balances Sheets.pptxCFO Pro+Analytics
 

More from CFO Pro+Analytics (17)

Managing a Loan Portfolio with Great Analytical Tools.pdf
Managing a Loan Portfolio with Great Analytical Tools.pdfManaging a Loan Portfolio with Great Analytical Tools.pdf
Managing a Loan Portfolio with Great Analytical Tools.pdf
 
Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...
Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...
Recurring Revenue Modeling Can be Tricky, Using Cancellation Curves Can Impro...
 
What's Really Going on With Consumer Balance Sheets?
What's Really Going on With Consumer Balance Sheets?What's Really Going on With Consumer Balance Sheets?
What's Really Going on With Consumer Balance Sheets?
 
Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...
Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...
Using a Survival Model for Credit Risk Scoring and Loan Pricing Instead of XG...
 
Applying Predictive Modeling to Crypto Futures Trading
Applying Predictive Modeling to Crypto Futures TradingApplying Predictive Modeling to Crypto Futures Trading
Applying Predictive Modeling to Crypto Futures Trading
 
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
 
Cracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptx
Cracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptxCracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptx
Cracks in Consumer Credit Card Delinquency Despite High Cash Balances.pptx
 
Consumer Credit Card Interest Savings in a Decreasing Rate Environment.pptx
Consumer Credit Card Interest Savings in a Decreasing Rate Environment.pptxConsumer Credit Card Interest Savings in a Decreasing Rate Environment.pptx
Consumer Credit Card Interest Savings in a Decreasing Rate Environment.pptx
 
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
How We Replaced an Implementation of Workday Adaptive Planning Enterprise Man...
 
Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...
Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...
Unveiling the Enigma Contrasting Consumer Cash Reserves with Escalating Credi...
 
Decoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptx
Decoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptxDecoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptx
Decoding Consumer Balance Sheets a Deeper Dive Beyond Savings Rates.pptx
 
How we replaced an implementation of workday adaptive planning enterprise man...
How we replaced an implementation of workday adaptive planning enterprise man...How we replaced an implementation of workday adaptive planning enterprise man...
How we replaced an implementation of workday adaptive planning enterprise man...
 
LCV-Understanding lifetime customer value-.pptx
LCV-Understanding lifetime customer value-.pptxLCV-Understanding lifetime customer value-.pptx
LCV-Understanding lifetime customer value-.pptx
 
Unlocking consumer loan pricing a deep dive into survival regression algorith...
Unlocking consumer loan pricing a deep dive into survival regression algorith...Unlocking consumer loan pricing a deep dive into survival regression algorith...
Unlocking consumer loan pricing a deep dive into survival regression algorith...
 
Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...
Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...
Unlocking Value Creation the Power of Lifetime Customer Value in Operational ...
 
Using a Survival Regression to Evaluate.pptx
Using a Survival Regression to Evaluate.pptxUsing a Survival Regression to Evaluate.pptx
Using a Survival Regression to Evaluate.pptx
 
Unraveling the Complexities of Consumer Balances Sheets.pptx
Unraveling the Complexities of Consumer Balances Sheets.pptxUnraveling the Complexities of Consumer Balances Sheets.pptx
Unraveling the Complexities of Consumer Balances Sheets.pptx
 

Recently uploaded

FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756dollysharma2066
 
Phases of Negotiation .pptx
 Phases of Negotiation .pptx Phases of Negotiation .pptx
Phases of Negotiation .pptxnandhinijagan9867
 
Cracking the Cultural Competence Code.pptx
Cracking the Cultural Competence Code.pptxCracking the Cultural Competence Code.pptx
Cracking the Cultural Competence Code.pptxWorkforce Group
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756dollysharma2066
 
Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000
Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000
Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000dlhescort
 
Falcon Invoice Discounting: Empowering Your Business Growth
Falcon Invoice Discounting: Empowering Your Business GrowthFalcon Invoice Discounting: Empowering Your Business Growth
Falcon Invoice Discounting: Empowering Your Business GrowthFalcon investment
 
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service AvailableCall Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service AvailableSeo
 
Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...
Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...
Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...Sheetaleventcompany
 
Business Model Canvas (BMC)- A new venture concept
Business Model Canvas (BMC)-  A new venture conceptBusiness Model Canvas (BMC)-  A new venture concept
Business Model Canvas (BMC)- A new venture conceptP&CO
 
It will be International Nurses' Day on 12 May
It will be International Nurses' Day on 12 MayIt will be International Nurses' Day on 12 May
It will be International Nurses' Day on 12 MayNZSG
 
Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...
Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...
Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...lizamodels9
 
Dr. Admir Softic_ presentation_Green Club_ENG.pdf
Dr. Admir Softic_ presentation_Green Club_ENG.pdfDr. Admir Softic_ presentation_Green Club_ENG.pdf
Dr. Admir Softic_ presentation_Green Club_ENG.pdfAdmir Softic
 
RSA Conference Exhibitor List 2024 - Exhibitors Data
RSA Conference Exhibitor List 2024 - Exhibitors DataRSA Conference Exhibitor List 2024 - Exhibitors Data
RSA Conference Exhibitor List 2024 - Exhibitors DataExhibitors Data
 
The Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai Kuwait
The Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai KuwaitThe Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai Kuwait
The Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai Kuwaitdaisycvs
 
Mysore Call Girls 8617370543 WhatsApp Number 24x7 Best Services
Mysore Call Girls 8617370543 WhatsApp Number 24x7 Best ServicesMysore Call Girls 8617370543 WhatsApp Number 24x7 Best Services
Mysore Call Girls 8617370543 WhatsApp Number 24x7 Best ServicesDipal Arora
 
Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...
Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...
Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...daisycvs
 
SEO Case Study: How I Increased SEO Traffic & Ranking by 50-60% in 6 Months
SEO Case Study: How I Increased SEO Traffic & Ranking by 50-60%  in 6 MonthsSEO Case Study: How I Increased SEO Traffic & Ranking by 50-60%  in 6 Months
SEO Case Study: How I Increased SEO Traffic & Ranking by 50-60% in 6 MonthsIndeedSEO
 
Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876
Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876
Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876dlhescort
 
Eluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort Service
Eluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort ServiceEluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort Service
Eluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort ServiceDamini Dixit
 

Recently uploaded (20)

FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Majnu Ka Tilla, Delhi Contact Us 8377877756
 
Phases of Negotiation .pptx
 Phases of Negotiation .pptx Phases of Negotiation .pptx
Phases of Negotiation .pptx
 
Cracking the Cultural Competence Code.pptx
Cracking the Cultural Competence Code.pptxCracking the Cultural Competence Code.pptx
Cracking the Cultural Competence Code.pptx
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
 
(Anamika) VIP Call Girls Napur Call Now 8617697112 Napur Escorts 24x7
(Anamika) VIP Call Girls Napur Call Now 8617697112 Napur Escorts 24x7(Anamika) VIP Call Girls Napur Call Now 8617697112 Napur Escorts 24x7
(Anamika) VIP Call Girls Napur Call Now 8617697112 Napur Escorts 24x7
 
Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000
Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000
Call Girls In Majnu Ka Tilla 959961~3876 Shot 2000 Night 8000
 
Falcon Invoice Discounting: Empowering Your Business Growth
Falcon Invoice Discounting: Empowering Your Business GrowthFalcon Invoice Discounting: Empowering Your Business Growth
Falcon Invoice Discounting: Empowering Your Business Growth
 
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service AvailableCall Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
 
Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...
Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...
Chandigarh Escorts Service 📞8868886958📞 Just📲 Call Nihal Chandigarh Call Girl...
 
Business Model Canvas (BMC)- A new venture concept
Business Model Canvas (BMC)-  A new venture conceptBusiness Model Canvas (BMC)-  A new venture concept
Business Model Canvas (BMC)- A new venture concept
 
It will be International Nurses' Day on 12 May
It will be International Nurses' Day on 12 MayIt will be International Nurses' Day on 12 May
It will be International Nurses' Day on 12 May
 
Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...
Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...
Call Girls From Pari Chowk Greater Noida ❤️8448577510 ⊹Best Escorts Service I...
 
Dr. Admir Softic_ presentation_Green Club_ENG.pdf
Dr. Admir Softic_ presentation_Green Club_ENG.pdfDr. Admir Softic_ presentation_Green Club_ENG.pdf
Dr. Admir Softic_ presentation_Green Club_ENG.pdf
 
RSA Conference Exhibitor List 2024 - Exhibitors Data
RSA Conference Exhibitor List 2024 - Exhibitors DataRSA Conference Exhibitor List 2024 - Exhibitors Data
RSA Conference Exhibitor List 2024 - Exhibitors Data
 
The Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai Kuwait
The Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai KuwaitThe Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai Kuwait
The Abortion pills for sale in Qatar@Doha [+27737758557] []Deira Dubai Kuwait
 
Mysore Call Girls 8617370543 WhatsApp Number 24x7 Best Services
Mysore Call Girls 8617370543 WhatsApp Number 24x7 Best ServicesMysore Call Girls 8617370543 WhatsApp Number 24x7 Best Services
Mysore Call Girls 8617370543 WhatsApp Number 24x7 Best Services
 
Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...
Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...
Quick Doctor In Kuwait +2773`7758`557 Kuwait Doha Qatar Dubai Abu Dhabi Sharj...
 
SEO Case Study: How I Increased SEO Traffic & Ranking by 50-60% in 6 Months
SEO Case Study: How I Increased SEO Traffic & Ranking by 50-60%  in 6 MonthsSEO Case Study: How I Increased SEO Traffic & Ranking by 50-60%  in 6 Months
SEO Case Study: How I Increased SEO Traffic & Ranking by 50-60% in 6 Months
 
Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876
Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876
Cheap Rate Call Girls In Noida Sector 62 Metro 959961乂3876
 
Eluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort Service
Eluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort ServiceEluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort Service
Eluru Call Girls Service ☎ ️93326-06886 ❤️‍🔥 Enjoy 24/7 Escort Service
 

A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model

  • 1. A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model Created by: Salvatore Tirabassi Document Copyright 2023
  • 2. A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model Our data science and analytics teams handle and apply lots of data for insightful decision-making. Last year, I presented the data science team with a challenge: use historical data to predict a key business driver for each of the next 8 periods. We wanted to have a data-driven preview of what we might see in the in each of the next eight periods so that we could anticipate the actual outcome and make better decisions with an eight-period window. Document Copyright 2023
  • 3. The data science team went to work researching ways we could do this and tested a few different methodologies. We had lots of input data from our own and public sources to feed any model we wanted to test, which worked well for us. With that said, we had low expectations about finding a predictive model that produced anything reliable. Testing different algorithms is always our approach. For the semi-technical readers, before settling on Temporal Fusion Transformer (TFT), the algorithms we tested included ARIMA, VAR, GARCH, ARCH models (univariate), Prophet, NHits, and Nbeats. TFT is an attention-based deep learning neural network algorithm. Using a mix of inputs, it produces a forecast over multiple periods in a future time horizon that you can determine. You can predict days, weeks, months quarters (really any interval is possible) into the future. Your choice. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model
  • 4. The picture below shows the concept of how TFT works. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model Source: Bryan Lim, Sercan Ö. Arık, Nicolas Loeff, Tomas Pfister. “Temporal Fusion Transformers for interpretable multi- horizon time series forecasting.”
  • 5. A continuous improvement process best describes how we developed and continue to refine the model. It’s a never ending process of improvement, as a true crystal ball is never achieved. These are the four stages of development we went through after choosing TFT as our algorithm: Stage 1a: Selecting all logical observed inputs and test how they drive the model. We started with over 100 and the final model only used 20. Go to Stage 1b as needed. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model
  • 6. Stage 1b: Refining the time intervals of the observed inputs. Since the inputs might come in varying time intervals (daily, weekly, monthly and quarterly, etc…), we needed to find methods to standardize them. You should choose an interval that matches the decision-making forecast you are producing, if you can. Go back to Stage 1a as needed. Stage 2: Model iteration and improvement. Complete back testing. Examine early predictions. Go back to Stages 1a and 1b as needed. At this point, you have probably settled on one or two of the most promising algorithms. Stage 3: Begin using in production and comparing predictions to the future periods as they unfold. Learn and refine by going back to any previous stage as needed. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model
  • 7. Stage 4: Continuous improvement loop. Write a long-term road map. Test new inputs as they are presented. Continuous scrutiny of the predictions against what actually happens – learn and make changes by going back to any previous stage as needed. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model Note that at any of the stages of development, you can use a TFT encoder decoder to measure the importance of different inputs in the algorithm to learn which ones have the most impact on your prediction.
  • 8. Below are the results of our model. The orange line is the actual result of the key driver and the blue line is the prediction of the key driver that was made 8 periods prior. The area to the right without the orange line is the next 8-period forecast. So, at Period 19, we use the blue line forecast to take action based on what Periods 20-27 tell us. When we reach Period 20, we evaluate the updated forecast and we make a decisions accordingly for the future periods. This way, we have a rolling 8-period prediction/decision cycle. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model
  • 9. Document Copyright 2023 A Machine Learning (ML) Crystal Ball? How We Predict Future Outcomes Using a Temporal Fusion Transformer Model As you can see the model has been refined to a level that it makes useful predictions and handles volatility of the prediction with some reliability. Right now, we don’t use this to predict the future down to the exact number, which would be ideal, but we use it for a directional understanding of where things are headed so we can make better decisions at