SlideShare a Scribd company logo
A REVIEW OF
NET LIFT MODELS
J U N E , 2 0 1 3 Z I X I A W A N G
S U M M E R Z W A N G @ G M A I L . C O M
BACKGROUND
• The true effectiveness of a market Champaign should be measured by the incremental
impact, which are the purchases that would not have taken place in absence of the
campaign rather than the gross number of purchase.
• Traditional propensity models (response model)
maximize the gross purchase rate
• Net lift models (incremental model, uplift model or true lift model)
Maximize the incremental impact/lift
Incremental Impact=Test group purchase rate (Gross purchase rate) - Control Group
Purchase Rate (self-selection purchase rate)
BACKGROUND
• There are four types of customers in the test and
control sets based on their response to the
marketing campaign.
• We can see that the traditional propensity models
are focus on all people that make purchases which
includes both self-selectors and swing clients but
net lift models will be able to identify swing clients
who are make the most incremental sale from you
marketing campaigns.
• Net models can significantly increase the net
impact of marketing campaigns when you have a
large number of self-selectors.
BASIC CONCEPT
• There are a few statistical concepts need to be known in order to understand the
fundamental of net lift model.
• Weight of evidence (WOE):
Describes the relationship (pattern) between a binary variable and a predictor
WOE>0 : positive impact
WOE=0: no impact
WOE<0: negative impact
Calculation methods:
Kernel density estimators: http://en.wikipedia.org/wiki/Kernel_density_estimation
Histogram estimator
BASIC CONCEPT
• Information value (IV):
Measure the strength of relationship
• WOE=ln(0.05/0.06)=-0.182 IV=-0.182*(0.05-0.06)=0.002
• Usually 0.02 or 0.05 are used as the cut point for the IV value to determine if the variable has significant
impact.
BASIC CONCEPT
• Penalized IV (PIV):
Measure the robustness of WOE and IV
For each bin, penalty is calculated as the difference of WOE between training and
validation sample.
Total Penalty= Sum (Penalty in each bin * (% responders-% non-responders))
Penalized IV= IV - Total Penalty. The suggested cutoff for PIV is 0.1 for variable
selection.
If the Total Penalty is relative small to IV then we can consider the variable is robust.
Only include variables that with relative large penalized IV in the final model.
• Net Weight of Evidence (NWOE):
NWOE=WOE(test)-WOE(control)
• Net Information Value (NIV):
Net IV describes the net strength.
• Penalized NIV:
measured the robustness of a variable.
MODELING METHOD
• Regression-based methods:
1) DSM (Different score models)
Method 1: Build two separate logistic regression models
Incremental lift score= P(purchase | treatment)-P(purchase | control)
Method 2: A single logistic regression model (the bifurcated logistic model)
Logit(P(reponse|X) = a + b*X + g*treatment + l* treatment *X
score = P(response|X,treatment =1) - P(response|X,treatment =0)
1) PDM (Probability decomposition models)
When the test and control group are equally sized:
P(purchase due to treatment)=P(purchase | treatment)*(2-1/P(treatment | purchase))
Otherwise :
P(purchase due to treatment)=P(purchase | treatment)*(1+Nt/Nc*1/P(treatment |
purchase))
MODELING METHOD
• Non-regression methods:
1) uplift Radom forest:
This method estimate personalized treatment effects by binary recursive partitioning. The
estimated personalized treatment effect is obtained by averaging the predictions of the
individual trees in the ensemble.
2) KNN( K-nearest-neighbors) classifiers
This method use the net purchase rate calculated from a nearest neighborhood of customers
form the training set to estimate the net score for observations in the validation dataset.
3) Net Naive ( and Semi-Naïve) Bayes classifier
Naive bayes classifier assumes that all predictors are conditionally independent given the
target variable Y. The net score using net naive Bayes method would just be the net weight
of evidence (NWOE). The generalized version Net naïve Bayes method rotated the WOE table
and make them more orthogonal to each other.
METHOD COMPARISON
• Regression-based Methods:
1) No attempt to maximize the incremental purchase rate directly
2) Subtracting two independent models can present a black box
3) Little control over the smoothness of the final prediction functions
• Non-regression methods:
1) Fitting the incremental purchase rate more directly
2) If NBC or SNBC methods are used, the prediction functions can be interpreted directly
and we can control the smoothness of these functions. For KNN, it's still a black box.
3) It is fitting an inherently unstable target( double variable) and can be over-fitting.
Therefore, full validation or forward validation are needed.
EVALUATING THE EFFECTIVENESS OF NET MODEL
• It’s still an area of on-going research.
• A commonly used way is to use the top two deciles or top 10% as a measure of
success.
• Based on the case example provided by Kim Larsen in 12th Annual data mining
conference, net difference score with bifurcated adaptive logistic regression works the
best followed by generalized net naive bayes, net naive bayes, KNN classifier and net
difference score with two linear logistic regressions.
• For your result, the clients have the high net score are swing clients describe in slices 3
and clients with 0 or even negative score are self-selector, no purchase or do not
disturb( sleeping dog).
IMPLEMENTATIONS
• A lot of examples that available online are build using the a series of macros coded in
SAS.
• There is uplift package for R using causal conditional inference trees to estimate
personalized treatment effects
• There are also R Package smbinning and R package WOE allow you to quickly
calculated the WOE and IV values.
APPLICATION
• Various of Marketing analysis:
Direct Mail
Email
Sweepstake
A/B testing
• There are different ways of to define the profit per purchase depends on your goal and
time windows. Measurements can be the net profit margin per sale, Net present value
(NPV), life time value (LTV).
• Net modeling has also been applied to personalized medicine.
REFERENCE
• Net Lift Models: Optimizing the Impact of Your Marketing Efforts by SAS institute
• Net Models presentation in 12th Annual data mining conference by Kim Larsen
https://www.youtube.com/watch?v=JN3WE8IZNVY
• Analyzing Collection effectiveness using Incremental Response Modeling by Ryan
Burton etc.
http://www.mwsug.org/proceedings/2014/BI/MWSUG-2014-BI06.pdf
• What are uplift models by Jeffrey Strickland
http://www.analyticbridge.com/profiles/blogs/what-are-uplift-models

More Related Content

What's hot

Introduction to Uplift Modelling
Introduction to Uplift ModellingIntroduction to Uplift Modelling
Introduction to Uplift Modelling
Pierre Gutierrez
 
Recommending What Video to Watch Next: A Multitask Ranking System
Recommending What Video to Watch Next: A Multitask Ranking SystemRecommending What Video to Watch Next: A Multitask Ranking System
Recommending What Video to Watch Next: A Multitask Ranking System
ivaderivader
 
Counterfactual evaluation of machine learning models
Counterfactual evaluation of machine learning modelsCounterfactual evaluation of machine learning models
Counterfactual evaluation of machine learning models
Michael Manapat
 
Feature Engineering for ML - Dmitry Larko, H2O.ai
Feature Engineering for ML - Dmitry Larko, H2O.aiFeature Engineering for ML - Dmitry Larko, H2O.ai
Feature Engineering for ML - Dmitry Larko, H2O.ai
Sri Ambati
 
KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...
KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...
KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...
Edureka!
 
Inspiring Alignment and Autonomy - The Leaders Role in Scaling Agile
Inspiring Alignment and Autonomy - The Leaders Role in Scaling AgileInspiring Alignment and Autonomy - The Leaders Role in Scaling Agile
Inspiring Alignment and Autonomy - The Leaders Role in Scaling Agile
Leland Newsom CSP-SM, SPC5, SDP
 
Knn
KnnKnn
Recommendation Systems - Why How and Real Life Applications
Recommendation Systems - Why How and Real Life ApplicationsRecommendation Systems - Why How and Real Life Applications
Recommendation Systems - Why How and Real Life Applications
Liron Zighelnic
 
Recommendations for Building Machine Learning Software
Recommendations for Building Machine Learning SoftwareRecommendations for Building Machine Learning Software
Recommendations for Building Machine Learning Software
Justin Basilico
 
Recommender Systems - A Review and Recent Research Trends
Recommender Systems  -  A Review and Recent Research TrendsRecommender Systems  -  A Review and Recent Research Trends
Recommender Systems - A Review and Recent Research Trends
Sujoy Bag
 
Past, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspectivePast, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspective
Xavier Amatriain
 
Past, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectivePast, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry Perspective
Justin Basilico
 
Active learning literature survey
Active learning  literature surveyActive learning  literature survey
Active learning literature survey
hyunsikkim30
 
Kaggle winning solutions: Retail Sales Forecasting
Kaggle winning solutions: Retail Sales ForecastingKaggle winning solutions: Retail Sales Forecasting
Kaggle winning solutions: Retail Sales Forecasting
Yan Xu
 
Matrix Factorization In Recommender Systems
Matrix Factorization In Recommender SystemsMatrix Factorization In Recommender Systems
Matrix Factorization In Recommender Systems
YONG ZHENG
 
Artwork Personalization at Netflix
Artwork Personalization at NetflixArtwork Personalization at Netflix
Artwork Personalization at Netflix
Justin Basilico
 
Recommender Systems
Recommender SystemsRecommender Systems
Recommender Systems
Francesco Casalegno
 
Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...
Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...
Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...
Kishor Datta Gupta
 
Why start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaignsWhy start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaigns
Data Con LA
 
K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...
K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...
K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...
Edureka!
 

What's hot (20)

Introduction to Uplift Modelling
Introduction to Uplift ModellingIntroduction to Uplift Modelling
Introduction to Uplift Modelling
 
Recommending What Video to Watch Next: A Multitask Ranking System
Recommending What Video to Watch Next: A Multitask Ranking SystemRecommending What Video to Watch Next: A Multitask Ranking System
Recommending What Video to Watch Next: A Multitask Ranking System
 
Counterfactual evaluation of machine learning models
Counterfactual evaluation of machine learning modelsCounterfactual evaluation of machine learning models
Counterfactual evaluation of machine learning models
 
Feature Engineering for ML - Dmitry Larko, H2O.ai
Feature Engineering for ML - Dmitry Larko, H2O.aiFeature Engineering for ML - Dmitry Larko, H2O.ai
Feature Engineering for ML - Dmitry Larko, H2O.ai
 
KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...
KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...
KNN Algorithm using Python | How KNN Algorithm works | Python Data Science Tr...
 
Inspiring Alignment and Autonomy - The Leaders Role in Scaling Agile
Inspiring Alignment and Autonomy - The Leaders Role in Scaling AgileInspiring Alignment and Autonomy - The Leaders Role in Scaling Agile
Inspiring Alignment and Autonomy - The Leaders Role in Scaling Agile
 
Knn
KnnKnn
Knn
 
Recommendation Systems - Why How and Real Life Applications
Recommendation Systems - Why How and Real Life ApplicationsRecommendation Systems - Why How and Real Life Applications
Recommendation Systems - Why How and Real Life Applications
 
Recommendations for Building Machine Learning Software
Recommendations for Building Machine Learning SoftwareRecommendations for Building Machine Learning Software
Recommendations for Building Machine Learning Software
 
Recommender Systems - A Review and Recent Research Trends
Recommender Systems  -  A Review and Recent Research TrendsRecommender Systems  -  A Review and Recent Research Trends
Recommender Systems - A Review and Recent Research Trends
 
Past, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspectivePast, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspective
 
Past, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectivePast, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry Perspective
 
Active learning literature survey
Active learning  literature surveyActive learning  literature survey
Active learning literature survey
 
Kaggle winning solutions: Retail Sales Forecasting
Kaggle winning solutions: Retail Sales ForecastingKaggle winning solutions: Retail Sales Forecasting
Kaggle winning solutions: Retail Sales Forecasting
 
Matrix Factorization In Recommender Systems
Matrix Factorization In Recommender SystemsMatrix Factorization In Recommender Systems
Matrix Factorization In Recommender Systems
 
Artwork Personalization at Netflix
Artwork Personalization at NetflixArtwork Personalization at Netflix
Artwork Personalization at Netflix
 
Recommender Systems
Recommender SystemsRecommender Systems
Recommender Systems
 
Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...
Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...
Deep Reinforcement Learning based Recommendation with Explicit User-ItemInter...
 
Why start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaignsWhy start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaigns
 
K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...
K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...
K-Means Clustering Algorithm - Cluster Analysis | Machine Learning Algorithm ...
 

Viewers also liked

5 simple questions to determin sample size
5 simple questions to determin sample size5 simple questions to determin sample size
5 simple questions to determin sample size
Zixia Wang
 
OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...
OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...
OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...
OpenSymmetry
 
OpenSymmetry - Business Intelligence Maturity
OpenSymmetry - Business Intelligence MaturityOpenSymmetry - Business Intelligence Maturity
OpenSymmetry - Business Intelligence Maturity
OpenSymmetry
 
Uplift Modeling: Optimize for Influence and Persuade by the Numbers
Uplift Modeling: Optimize for Influence and Persuade by the NumbersUplift Modeling: Optimize for Influence and Persuade by the Numbers
Uplift Modeling: Optimize for Influence and Persuade by the Numbers
Rising Media Ltd.
 
From Labelling Open data images to building a private recommender system
From Labelling Open data images to building a private recommender systemFrom Labelling Open data images to building a private recommender system
From Labelling Open data images to building a private recommender system
Pierre Gutierrez
 
Churn prediction data modeling
Churn prediction data modelingChurn prediction data modeling
Churn prediction data modeling
Pierre Gutierrez
 

Viewers also liked (6)

5 simple questions to determin sample size
5 simple questions to determin sample size5 simple questions to determin sample size
5 simple questions to determin sample size
 
OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...
OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...
OpenSymmetry WorldatWork 2013 Conference Workshop - Sales Compensation Best P...
 
OpenSymmetry - Business Intelligence Maturity
OpenSymmetry - Business Intelligence MaturityOpenSymmetry - Business Intelligence Maturity
OpenSymmetry - Business Intelligence Maturity
 
Uplift Modeling: Optimize for Influence and Persuade by the Numbers
Uplift Modeling: Optimize for Influence and Persuade by the NumbersUplift Modeling: Optimize for Influence and Persuade by the Numbers
Uplift Modeling: Optimize for Influence and Persuade by the Numbers
 
From Labelling Open data images to building a private recommender system
From Labelling Open data images to building a private recommender systemFrom Labelling Open data images to building a private recommender system
From Labelling Open data images to building a private recommender system
 
Churn prediction data modeling
Churn prediction data modelingChurn prediction data modeling
Churn prediction data modeling
 

Similar to A review of net lift models

Vi sem
Vi semVi sem
Satisfaction and loyalty
Satisfaction and loyaltySatisfaction and loyalty
Satisfaction and loyalty
TheDataNation
 
Predictive Analytics for Customer Targeting: A Telemarketing Banking Example
Predictive Analytics for Customer Targeting: A Telemarketing Banking ExamplePredictive Analytics for Customer Targeting: A Telemarketing Banking Example
Predictive Analytics for Customer Targeting: A Telemarketing Banking Example
Pedro Ecija Serrano
 
Credit scorecard
Credit scorecardCredit scorecard
Credit scorecard
Tuhin AI Advisory
 
IRJET- Finding Optimal Skyline Product Combinations Under Price Promotion
IRJET- Finding Optimal Skyline Product Combinations Under Price PromotionIRJET- Finding Optimal Skyline Product Combinations Under Price Promotion
IRJET- Finding Optimal Skyline Product Combinations Under Price Promotion
IRJET Journal
 
Statistical Learning and Model Selection (1).pptx
Statistical Learning and Model Selection (1).pptxStatistical Learning and Model Selection (1).pptx
Statistical Learning and Model Selection (1).pptx
rajalakshmi5921
 
Insurance Churn Prediction Data Analysis Project
Insurance Churn Prediction Data Analysis ProjectInsurance Churn Prediction Data Analysis Project
Insurance Churn Prediction Data Analysis Project
Boston Institute of Analytics
 
Captsone_Paper_Alexander
Captsone_Paper_AlexanderCaptsone_Paper_Alexander
Captsone_Paper_Alexander
Alexander Barriga, M.S.
 
SLIDESHARE PPT-DECISION MAKING METHODS.pptx
SLIDESHARE PPT-DECISION MAKING METHODS.pptxSLIDESHARE PPT-DECISION MAKING METHODS.pptx
SLIDESHARE PPT-DECISION MAKING METHODS.pptx
CHAIRMAN M
 
Understanding statistics in laboratory quality control
Understanding statistics in laboratory quality controlUnderstanding statistics in laboratory quality control
Understanding statistics in laboratory quality control
Randox
 
Bank Customer Churn Prediction- Saurav Singh.pptx
Bank Customer Churn Prediction- Saurav Singh.pptxBank Customer Churn Prediction- Saurav Singh.pptx
Bank Customer Churn Prediction- Saurav Singh.pptx
Boston Institute of Analytics
 
Lead Scoring Group Case Study Presentation.pdf
Lead Scoring Group Case Study Presentation.pdfLead Scoring Group Case Study Presentation.pdf
Lead Scoring Group Case Study Presentation.pdf
KrishP2
 
Case study of s&amp;p 500
Case study of s&amp;p 500Case study of s&amp;p 500
Case study of s&amp;p 500
Professional Training Academy
 
IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...
IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...
IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...
The Risk, Banking and Finance Society
 
Stochastic Loss Reserving-General Insurance
Stochastic Loss Reserving-General InsuranceStochastic Loss Reserving-General Insurance
Stochastic Loss Reserving-General Insurance
Syed Danish Ali
 
Cmt learning objective 36 case study of s&amp;p 500
Cmt learning objective 36   case study of s&amp;p 500Cmt learning objective 36   case study of s&amp;p 500
Cmt learning objective 36 case study of s&amp;p 500
Professional Training Academy
 
segmentda
segmentdasegmentda
IRJET- Fatigue Analysis of Offshore Steel Structures
IRJET- Fatigue Analysis of Offshore Steel StructuresIRJET- Fatigue Analysis of Offshore Steel Structures
IRJET- Fatigue Analysis of Offshore Steel Structures
IRJET Journal
 
IRJET- Web based Hybrid Book Recommender System using Genetic Algorithm
IRJET- Web based Hybrid Book Recommender System using Genetic AlgorithmIRJET- Web based Hybrid Book Recommender System using Genetic Algorithm
IRJET- Web based Hybrid Book Recommender System using Genetic Algorithm
IRJET Journal
 
Introduction to machine learning
Introduction to machine learningIntroduction to machine learning
Introduction to machine learning
Sanghamitra Deb
 

Similar to A review of net lift models (20)

Vi sem
Vi semVi sem
Vi sem
 
Satisfaction and loyalty
Satisfaction and loyaltySatisfaction and loyalty
Satisfaction and loyalty
 
Predictive Analytics for Customer Targeting: A Telemarketing Banking Example
Predictive Analytics for Customer Targeting: A Telemarketing Banking ExamplePredictive Analytics for Customer Targeting: A Telemarketing Banking Example
Predictive Analytics for Customer Targeting: A Telemarketing Banking Example
 
Credit scorecard
Credit scorecardCredit scorecard
Credit scorecard
 
IRJET- Finding Optimal Skyline Product Combinations Under Price Promotion
IRJET- Finding Optimal Skyline Product Combinations Under Price PromotionIRJET- Finding Optimal Skyline Product Combinations Under Price Promotion
IRJET- Finding Optimal Skyline Product Combinations Under Price Promotion
 
Statistical Learning and Model Selection (1).pptx
Statistical Learning and Model Selection (1).pptxStatistical Learning and Model Selection (1).pptx
Statistical Learning and Model Selection (1).pptx
 
Insurance Churn Prediction Data Analysis Project
Insurance Churn Prediction Data Analysis ProjectInsurance Churn Prediction Data Analysis Project
Insurance Churn Prediction Data Analysis Project
 
Captsone_Paper_Alexander
Captsone_Paper_AlexanderCaptsone_Paper_Alexander
Captsone_Paper_Alexander
 
SLIDESHARE PPT-DECISION MAKING METHODS.pptx
SLIDESHARE PPT-DECISION MAKING METHODS.pptxSLIDESHARE PPT-DECISION MAKING METHODS.pptx
SLIDESHARE PPT-DECISION MAKING METHODS.pptx
 
Understanding statistics in laboratory quality control
Understanding statistics in laboratory quality controlUnderstanding statistics in laboratory quality control
Understanding statistics in laboratory quality control
 
Bank Customer Churn Prediction- Saurav Singh.pptx
Bank Customer Churn Prediction- Saurav Singh.pptxBank Customer Churn Prediction- Saurav Singh.pptx
Bank Customer Churn Prediction- Saurav Singh.pptx
 
Lead Scoring Group Case Study Presentation.pdf
Lead Scoring Group Case Study Presentation.pdfLead Scoring Group Case Study Presentation.pdf
Lead Scoring Group Case Study Presentation.pdf
 
Case study of s&amp;p 500
Case study of s&amp;p 500Case study of s&amp;p 500
Case study of s&amp;p 500
 
IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...
IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...
IRMC2016- Keynote Speech - Giovanni Barone Adesi - Lecture title: “Crude Oil ...
 
Stochastic Loss Reserving-General Insurance
Stochastic Loss Reserving-General InsuranceStochastic Loss Reserving-General Insurance
Stochastic Loss Reserving-General Insurance
 
Cmt learning objective 36 case study of s&amp;p 500
Cmt learning objective 36   case study of s&amp;p 500Cmt learning objective 36   case study of s&amp;p 500
Cmt learning objective 36 case study of s&amp;p 500
 
segmentda
segmentdasegmentda
segmentda
 
IRJET- Fatigue Analysis of Offshore Steel Structures
IRJET- Fatigue Analysis of Offshore Steel StructuresIRJET- Fatigue Analysis of Offshore Steel Structures
IRJET- Fatigue Analysis of Offshore Steel Structures
 
IRJET- Web based Hybrid Book Recommender System using Genetic Algorithm
IRJET- Web based Hybrid Book Recommender System using Genetic AlgorithmIRJET- Web based Hybrid Book Recommender System using Genetic Algorithm
IRJET- Web based Hybrid Book Recommender System using Genetic Algorithm
 
Introduction to machine learning
Introduction to machine learningIntroduction to machine learning
Introduction to machine learning
 

Recently uploaded

Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...
Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...
Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...
Kaxil Naik
 
原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理
原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理 原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理
原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理
tzu5xla
 
UofT毕业证如何办理
UofT毕业证如何办理UofT毕业证如何办理
UofT毕业证如何办理
exukyp
 
How To Control IO Usage using Resource Manager
How To Control IO Usage using Resource ManagerHow To Control IO Usage using Resource Manager
How To Control IO Usage using Resource Manager
Alireza Kamrani
 
writing report business partner b1+ .pdf
writing report business partner b1+ .pdfwriting report business partner b1+ .pdf
writing report business partner b1+ .pdf
VyNguyen709676
 
一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理
一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理
一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理
lzdvtmy8
 
University of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma TranscriptUniversity of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma Transcript
soxrziqu
 
一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理
nuttdpt
 
一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理
一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理
一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理
nuttdpt
 
一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理
一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理
一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理
ywqeos
 
DSSML24_tspann_CodelessGenerativeAIPipelines
DSSML24_tspann_CodelessGenerativeAIPipelinesDSSML24_tspann_CodelessGenerativeAIPipelines
DSSML24_tspann_CodelessGenerativeAIPipelines
Timothy Spann
 
一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理
一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理
一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理
asyed10
 
Sample Devops SRE Product Companies .pdf
Sample Devops SRE  Product Companies .pdfSample Devops SRE  Product Companies .pdf
Sample Devops SRE Product Companies .pdf
Vineet
 
Jio cinema Retention & Engagement Strategy.pdf
Jio cinema Retention & Engagement Strategy.pdfJio cinema Retention & Engagement Strategy.pdf
Jio cinema Retention & Engagement Strategy.pdf
inaya7568
 
一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理
一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理
一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理
slg6lamcq
 
一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理
一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理
一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理
1tyxnjpia
 
Building a Quantum Computer Neutral Atom.pdf
Building a Quantum Computer Neutral Atom.pdfBuilding a Quantum Computer Neutral Atom.pdf
Building a Quantum Computer Neutral Atom.pdf
cjimenez2581
 
一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理
bmucuha
 
一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理
一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理
一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理
nyfuhyz
 
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
taqyea
 

Recently uploaded (20)

Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...
Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...
Orchestrating the Future: Navigating Today's Data Workflow Challenges with Ai...
 
原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理
原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理 原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理
原版一比一爱尔兰都柏林大学毕业证(UCD毕业证书)如何办理
 
UofT毕业证如何办理
UofT毕业证如何办理UofT毕业证如何办理
UofT毕业证如何办理
 
How To Control IO Usage using Resource Manager
How To Control IO Usage using Resource ManagerHow To Control IO Usage using Resource Manager
How To Control IO Usage using Resource Manager
 
writing report business partner b1+ .pdf
writing report business partner b1+ .pdfwriting report business partner b1+ .pdf
writing report business partner b1+ .pdf
 
一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理
一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理
一比一原版格里菲斯大学毕业证(Griffith毕业证书)学历如何办理
 
University of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma TranscriptUniversity of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma Transcript
 
一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB文凭证书)圣芭芭拉分校毕业证如何办理
 
一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理
一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理
一比一原版(UCSF文凭证书)旧金山分校毕业证如何办理
 
一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理
一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理
一比一原版(lbs毕业证书)伦敦商学院毕业证如何办理
 
DSSML24_tspann_CodelessGenerativeAIPipelines
DSSML24_tspann_CodelessGenerativeAIPipelinesDSSML24_tspann_CodelessGenerativeAIPipelines
DSSML24_tspann_CodelessGenerativeAIPipelines
 
一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理
一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理
一比一原版美国帕森斯设计学院毕业证(parsons毕业证书)如何办理
 
Sample Devops SRE Product Companies .pdf
Sample Devops SRE  Product Companies .pdfSample Devops SRE  Product Companies .pdf
Sample Devops SRE Product Companies .pdf
 
Jio cinema Retention & Engagement Strategy.pdf
Jio cinema Retention & Engagement Strategy.pdfJio cinema Retention & Engagement Strategy.pdf
Jio cinema Retention & Engagement Strategy.pdf
 
一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理
一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理
一比一原版南十字星大学毕业证(SCU毕业证书)学历如何办理
 
一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理
一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理
一比一原版(Sheffield毕业证书)谢菲尔德大学毕业证如何办理
 
Building a Quantum Computer Neutral Atom.pdf
Building a Quantum Computer Neutral Atom.pdfBuilding a Quantum Computer Neutral Atom.pdf
Building a Quantum Computer Neutral Atom.pdf
 
一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理
 
一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理
一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理
一比一原版(UMN文凭证书)明尼苏达大学毕业证如何办理
 
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
 

A review of net lift models

  • 1. A REVIEW OF NET LIFT MODELS J U N E , 2 0 1 3 Z I X I A W A N G S U M M E R Z W A N G @ G M A I L . C O M
  • 2. BACKGROUND • The true effectiveness of a market Champaign should be measured by the incremental impact, which are the purchases that would not have taken place in absence of the campaign rather than the gross number of purchase. • Traditional propensity models (response model) maximize the gross purchase rate • Net lift models (incremental model, uplift model or true lift model) Maximize the incremental impact/lift Incremental Impact=Test group purchase rate (Gross purchase rate) - Control Group Purchase Rate (self-selection purchase rate)
  • 3. BACKGROUND • There are four types of customers in the test and control sets based on their response to the marketing campaign. • We can see that the traditional propensity models are focus on all people that make purchases which includes both self-selectors and swing clients but net lift models will be able to identify swing clients who are make the most incremental sale from you marketing campaigns. • Net models can significantly increase the net impact of marketing campaigns when you have a large number of self-selectors.
  • 4. BASIC CONCEPT • There are a few statistical concepts need to be known in order to understand the fundamental of net lift model. • Weight of evidence (WOE): Describes the relationship (pattern) between a binary variable and a predictor WOE>0 : positive impact WOE=0: no impact WOE<0: negative impact Calculation methods: Kernel density estimators: http://en.wikipedia.org/wiki/Kernel_density_estimation Histogram estimator
  • 5. BASIC CONCEPT • Information value (IV): Measure the strength of relationship • WOE=ln(0.05/0.06)=-0.182 IV=-0.182*(0.05-0.06)=0.002 • Usually 0.02 or 0.05 are used as the cut point for the IV value to determine if the variable has significant impact.
  • 6. BASIC CONCEPT • Penalized IV (PIV): Measure the robustness of WOE and IV For each bin, penalty is calculated as the difference of WOE between training and validation sample. Total Penalty= Sum (Penalty in each bin * (% responders-% non-responders)) Penalized IV= IV - Total Penalty. The suggested cutoff for PIV is 0.1 for variable selection. If the Total Penalty is relative small to IV then we can consider the variable is robust. Only include variables that with relative large penalized IV in the final model. • Net Weight of Evidence (NWOE): NWOE=WOE(test)-WOE(control) • Net Information Value (NIV): Net IV describes the net strength. • Penalized NIV: measured the robustness of a variable.
  • 7. MODELING METHOD • Regression-based methods: 1) DSM (Different score models) Method 1: Build two separate logistic regression models Incremental lift score= P(purchase | treatment)-P(purchase | control) Method 2: A single logistic regression model (the bifurcated logistic model) Logit(P(reponse|X) = a + b*X + g*treatment + l* treatment *X score = P(response|X,treatment =1) - P(response|X,treatment =0) 1) PDM (Probability decomposition models) When the test and control group are equally sized: P(purchase due to treatment)=P(purchase | treatment)*(2-1/P(treatment | purchase)) Otherwise : P(purchase due to treatment)=P(purchase | treatment)*(1+Nt/Nc*1/P(treatment | purchase))
  • 8. MODELING METHOD • Non-regression methods: 1) uplift Radom forest: This method estimate personalized treatment effects by binary recursive partitioning. The estimated personalized treatment effect is obtained by averaging the predictions of the individual trees in the ensemble. 2) KNN( K-nearest-neighbors) classifiers This method use the net purchase rate calculated from a nearest neighborhood of customers form the training set to estimate the net score for observations in the validation dataset. 3) Net Naive ( and Semi-Naïve) Bayes classifier Naive bayes classifier assumes that all predictors are conditionally independent given the target variable Y. The net score using net naive Bayes method would just be the net weight of evidence (NWOE). The generalized version Net naïve Bayes method rotated the WOE table and make them more orthogonal to each other.
  • 9. METHOD COMPARISON • Regression-based Methods: 1) No attempt to maximize the incremental purchase rate directly 2) Subtracting two independent models can present a black box 3) Little control over the smoothness of the final prediction functions • Non-regression methods: 1) Fitting the incremental purchase rate more directly 2) If NBC or SNBC methods are used, the prediction functions can be interpreted directly and we can control the smoothness of these functions. For KNN, it's still a black box. 3) It is fitting an inherently unstable target( double variable) and can be over-fitting. Therefore, full validation or forward validation are needed.
  • 10. EVALUATING THE EFFECTIVENESS OF NET MODEL • It’s still an area of on-going research. • A commonly used way is to use the top two deciles or top 10% as a measure of success. • Based on the case example provided by Kim Larsen in 12th Annual data mining conference, net difference score with bifurcated adaptive logistic regression works the best followed by generalized net naive bayes, net naive bayes, KNN classifier and net difference score with two linear logistic regressions. • For your result, the clients have the high net score are swing clients describe in slices 3 and clients with 0 or even negative score are self-selector, no purchase or do not disturb( sleeping dog).
  • 11. IMPLEMENTATIONS • A lot of examples that available online are build using the a series of macros coded in SAS. • There is uplift package for R using causal conditional inference trees to estimate personalized treatment effects • There are also R Package smbinning and R package WOE allow you to quickly calculated the WOE and IV values.
  • 12. APPLICATION • Various of Marketing analysis: Direct Mail Email Sweepstake A/B testing • There are different ways of to define the profit per purchase depends on your goal and time windows. Measurements can be the net profit margin per sale, Net present value (NPV), life time value (LTV). • Net modeling has also been applied to personalized medicine.
  • 13. REFERENCE • Net Lift Models: Optimizing the Impact of Your Marketing Efforts by SAS institute • Net Models presentation in 12th Annual data mining conference by Kim Larsen https://www.youtube.com/watch?v=JN3WE8IZNVY • Analyzing Collection effectiveness using Incremental Response Modeling by Ryan Burton etc. http://www.mwsug.org/proceedings/2014/BI/MWSUG-2014-BI06.pdf • What are uplift models by Jeffrey Strickland http://www.analyticbridge.com/profiles/blogs/what-are-uplift-models