R for Marketing Research and Analytics by Chris Chapman & Elea McDonnell Feit

Author:Chris Chapman & Elea McDonnell Feit
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

(9.1)

In this equation, the outcome of interest is y, and we compute its likelihood p(y) as a function of v x . We typically estimate v x as a function of the features (x) of a product, such as price. v x can take any real value, so we are able to treat it as a continuous function in a linear model. In that case, v x is composed from one or more coefficients of the model and indicates the importance of the corresponding features of the product.

This formula gives a value between [0, 1]. The likelihood of y is less than 50 % when v x is negative, is 50 % when v x  = 0, and is above 50 % when v x is positive. We compute this first by hand, and then switch to the equivalent, built-in plogis() function:

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.