TresVista - Card-Offer-Promotion

Chi Square Test Results

The data given is of credit records of individuals with certain attributes. Please go through following to understand the variables involved:

Var-Var Matrix

The below matrix represented in the following format: df, X-squared, p-value where,

Var/Var demographic_slice country_reg ad_exp card_offer
demographic_slice . 3 , 3.2525 , 0.3543 3 , 7.6743 , 0.05325 3 , 487.11 , < 2.2e-16
country_reg 3 , 3.2525 , 0.3543 . 1 , 0.010034 , 0.9202 1 , 176.85 , < 2.2e-16
ad_exp 3 , 7.6743 , 0.05325 1 , 0.010034 , 0.9202 . 1 , 0.018046 , 0.8931
card_offer 3 , 487.11 , < 2.2e-16 1 , 176.85 , < 2.2e-16 1 , 0.018046 , 0.8931 .

Confusion Matrix Statistics

For details and definition refer here

Parameters RandomForest BAGGING GBM C50 logisticReg Naive bayes
Accuracy 0.9824 0.978 0.9832 0.9848 0.9684 0.8896
95% CI 0.9764 , 0.9872 0.9714 , 0.9834 0.9773 , 0.9879 0.9792 , 0.9892 0.9608 , 0.9749 0.8766 , 0.9016
No Information Rate 0.8471 0.8471 0.8471 0.8471 0.8471 0.8471
P-Value (Acc > NIR) <2e-16 <2e-16 < 2e-16 < 2e-16 < 2e-16 4.93e-10
Kappa 0.9313 0.9144 0.9343 0.941 0.8759 0.3942
Mcnemar’s Test P-Value 0.1748 0.4185 0.08963 0.6265 0.1152 < 2.2e-16
Sensitivity 0.9920 0.9887 0.9929 0.9920 0.9849 1.0000
Specificity 0.9293 0.9188 0.9293 0.9450 0.8770 0.2775
Pos Pred Value 0.9873 0.9854 0.9873 0.9901 0.9780 0.8847
Neg Pred Value 0.9543 0.9360 0.9595 0.9550 0.9128 1.0000
Prevalence 0.8471 0.8471 0.8471 0.8471 0.8471 0.8471
Detection Rate 0.8403 0.8375 0.8411 0.8403 0.8343 0.8471
Detection Prevalence 0.8511 0.8499 0.8519 0.8487 0.8531 0.9576
Balanced Accuracy 0.9606 0.9538 0.9611 0.9685 0.9309 0.6387
ROC – AUC Results 0.9978 0.9955 0.9983 0.9988 0.994 0.9906
‘Positive’ Class No No No No No No

Evaluation Metric