|
Abstract:
|
While many years ago, it was difficult finding marketers that understood the value of incorporating predictive models into marketing campaigns, today most managers would agree that statistical models provide significant advantages. Indeed, this fact is so well accepted that even non-quantitative marketers have been taught how to assess the results of a modeling exercise. While this, in itself, is certainly an encouraging sign, model evaluation is not as simple a matter as many would believe. Many ingredients of the evaluation process are overlooked. And they can make a big difference in selecting the eventual model to deploy. Few marketers pose this important question; "what performance criteria did you use in selecting the final model?" Some believe, and to an extent correctly, that with the presence of a decile or performance report, the evaluation criteria are obvious.
|
