If you have a comment regarding a blog post, or wish to suggest a topic that you would like me to consider blogging on, then send me a message here. ¬†Thanks…

6 thoughts on “Contact”

  1. Can you comment on how to go about selection of stress testing models? Since the Fed variables are macroeconomic variables and represent time series, what types of models are best suited to highlight the sensitivity to the independent Fed variables when trying to explain delinquency rates, charge offs, reduction in borrower income on commercial loans? Thanks in advance for your help.

    1. You need to do historical regression fits of their macroeconomic data against your own operations, then use statistics to see what fits the best. MAPE is a favorite for determining fit, better than R-squared.

  2. Hi Ben. I just wanted to findout something. I have been working on an index covering the years 2009 to 2014 using normalized data. My problem now is that I want to add 2015 to the index. Since normalization uses the maximum and the minimum values of a series, if I add the observations for 2015, all my index change. how do I maintain the previous period index values e.g. 0.56 for 2013 and 0.48 for 2014? Your help will be greatly appreciated.


    1. Since based on a range, you will need to determine the new range (Xmax – Xmin), then ratio this against the old range (New/Old), then multiply the old values by this ratio, then use your new range going forward. This won’t be perfect though because your numerator Xi-Xmin won’t change, but it is a good approximation.

      With computers and spreadsheets, the real answer is to build your system to recalculate everything, that is the only way you will get a correct answer.

Leave a Reply

Your email address will not be published. Required fields are marked *