Posts

Showing posts with the label Multicollinearity

Predicting Life stage of a customer using the retail data

It is desirable for a retailer to have information of the life cycle stage of  a customer. This helps greatly to a retailer in offer optimizations. Targeting right customer with rights offers is greatest challenge for a retailer. They run hundreds of promotions and optimizing it saves huge cost and helps in retaining customers.  In the era of super markets, large retailers issue loyalty cards and maintain customer purchase data. But most of the demographic data is not available with them as customers normally do not fill these fields and leave them blank. Although customer demographic data are not available using purchase data certain life stages of customer can be predicted using rule base techniques or predictive modeling. Life stages bring changes in purchasing pattern. 1.  A single male will mostly buy his needs and daily eatables.  Eating habits gets changed if he is married etc.  A man can occasionally buy woman's product but it will be regular if ...

Multicollinearity

Multicollinearity is defined as the linear relationship between two or more independent variables while performing regression analysis between a dependent variables and set of independent variables. Multicollinearity presents a severe problem during regression modelling. Inclusion of independent variables having linear relationship with each other leads to parameter estimation with higher standard error. This in turn leads to inaccurate parameter estimation. Furthermore, due to inaccurate parameters regression model becomes unstable. The unstable models performs badly on the validation and test samples. When model is unstable, its performance deteriorate very fast compared to stable model over the period, though model is scored on the data of sample of same population. In such a situation an analyst must investigate for the multicollinearity, before finalizing the model. Next question is how to investigate and which variable should be kept if some variables are found to ...