Choosing between models - a business view

We discussed how we use the AIC to choose between models. The standard definition of the AIC is:

AIC = -2 * log-likelihood + 2 * number of parameters

However, this is a statistician's view of a model, where the only criterion for including a parameter is whether it is statistically significant. A business view might be different, as each extra parameter in a system will cost you money. IT systems have to be specified, programmed, tested and maintained, for example, and IT staff are not cheap. Each extra parameter might therefore cost you 5,000 in development costs (say), so you might be inclined to only include parameters if they are really significant. One way of doing this is to increase the penalty for the number of parameters in the definition of the AIC as follows:

AIC = -2 * log-likelihood + N * number of parameters

where N can be 2 for a statistical view, as before. However, if you felt that complexity was expensive, you can be particularly demanding of extra parameters and set N to 3, 4 or 5 (say). Thus, the AIC is not only a tool for the statistician, but it can also be adapted to help make business decisions.

 

Comments

captcha

Find by key-word


RECENT POSTS

The Institute and Faculty of Actuaries in the UK has ... Read more
The resurgence of measles in Europe signals something of a ... Read more
When fitting a statistical model we want two things as ... Read more
Stephen Richards
Stephen Richards is the Managing Director of Longevitas
AIC options in Longevitas
Longevitas users can set the AIC multiplier in the Modelling area of the Configuration settings. The AIC multipler can be controlled from the Calculation Core section.