Simple questions deserve simple answers, not glib gibberish.  Businesspeople deserve to thoroughly understand the statistics they rely upon by having the methodologies explained in a familiar language and not be asked to accept anything on the faith that their analysts understand what they do not.  And beware the fancy and complicated.  It need not be.


A Businessperson’s Dollars and Cents Approach to Statistical “Value of Information”

If only one attribute were observable at underwriting, which one would you like it to be? 
The academic statistician's natural impulse is to build an elaborate model using as much data as is available. But the relative dollars and cents importance of each predictor is often lost in the impressive complexity. 

We at have evolved a rigorous yet common sense approach that helps the business analyst maintain focus on the most important things to monitor in a portfolio.  A working paper outlining our approach is available HERE.


 A Manager's Guide Through Random Forests




Leo Breiman (1928-2005)



Many of today's most acclaimed techniques for advanced statistical analysis are much easier to understand than the professional statistician would have us believe.

Using the intuitive (and cleverly named) Random Forests technique as an example, we trace its academic pedigree in the PowerPoint presentation HERE using as a guide the life's work of the technique's inventor, Leo Breiman.

Knowing exactly how a methodology works helps to demystify it and allows a business manager to assess when (if ever) a complex technique may be appropriate to his or her data analysis task.