evaluation of prediction models for marketing campaigns author: saharon rosset advisor: dr. hsu...

19
Evaluation of Prediction Models for Marketing Campaigns Author: Saharon R osset Advisor: Dr. Hsu Gradua te: Lin Yan-Cheng

Upload: morgan-stewart

Post on 01-Jan-2016

216 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Evaluation of Prediction Models for Marketing Campaigns

Author: Saharon RossetAdvisor: Dr. Hsu Graduate: Lin Yan-Cheng

Page 2: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Abstract

Discuss model-evaluation criteria about their robustness

Ex. Response Rate in Customer retention

Page 3: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Agenda

Introduction Model Evaluation

– Planning Campaigns– Performance Measures

Prediction Model Performance– From Sample to Population– Confidence Intervals

Case Study Conclusion Opinion

Page 4: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Motivation

dealing with marketing applications the issue of evaluating prediction models is following twofold– Evaluation has to be statistically sound– Evaluate models should utilize from business

perspective

Page 5: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Objective

To discuss some applicable model-evaluation and selection criteria

Page 6: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Model Evaluation

Evaluate the models’ performance on an independent test set

Adjust the models’ score to fit the full population distribution, in case it is expected to be different from the sample distribution used for training and test

Page 7: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Planning Campaigns

To measure by the amount of responders captured within the targeted population

The amount can be measured in two diff. way– Lift: How much better are we doing by using our

model to select the target population relative to a random selection of the target population

– RR: How frequently do we expect to encounter a responder when running our campaign?

Page 8: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Performance Measures

A, B : Total number of responders and non-responders, respectively

Aj, Bj: Total number of responders and non-responders, respectively, in the j-th top quantile.

j*(A+B)or(Aj+Bj): all cases in the j-th top quantile

A/(A+B): overall response rate

Page 9: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Measures at Pre-Specified Cutoff Points

Response Rate– RR(j) = Aj/(Aj+Bj)

Lift

Response Non-Response Ratio– RNR(j) = (Aj/A)/(Bj/B)

Page 10: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Comarison of Cut-Point Measures

Page 11: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Predicting Model Performance

Performance measures are usually calculated on a test sample data set

These measures need to be adjusted to the full population

Page 12: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

From Sample to Population

A, B : the number of responders and non-responders in the FP (full population), respectively.

a, b : the number of responders and non-responders in the TS (Test Set), respectively.

ai, bi :the number of responders and non-responders in percentile i in the TS

Page 13: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Transformation

Extrapolate each percentile pair( ai, bi) in the TS to (Ai, Bi) in the FP

Ai = ai (A / a)

(Ai, Bi) does not add up to a FP percentile, TS percentiles are merged or split in order to attain FP percentiles

Page 14: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Confidence Intervals

Percentile point-estimators are not sufficient for evaluating the model predictive ability

confidence intervals for predict a model’s performance on future data

Page 15: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Case Study

Amdocs is a leading provider of CRM, Billing and Order Management solutions to the communications and IP industry worldwide

Consider a prediction model for a retention campaign, in which responders are potential churners and the overall response rate is the overall churn rate

Page 16: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Legacy model vs. New model

Initially legacy models RR at 10% was 2.75 times better than new model, but that was evaluated based on different test populations. Churn rate is 4.5 times in legacy models

Page 17: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

RR vs. Lift vs. RNR

Page 18: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Conclusion

Discuss a few model-evaluation criteria about their robustness under changing population distributions

RR is a non-robust measure, Lift and RNR measures be commended to be used

Page 19: Evaluation of Prediction Models for Marketing Campaigns Author: Saharon Rosset Advisor: Dr. Hsu Graduate: Lin Yan-Cheng

Opinion

We need to consider the robustness of measure in our case before we conclude that.