It is surprising to read about model risk from an economist :-) but I agree with the fact that many people who question models, especially robust models based on a few but useful parameters, do that in the interest of exemplifying assumptions and improving the model. Such people would usually claim that the model is too simple or unreliable for the sake of it and until the next stage of evolution in their thinking.
However, there are examples of mis-trust within the discipline.
I was surprised to learn that Bank of England uses several models with different theoretical foundations in order to forecast inflation and then provides the final estimate as a weighted average of outputs of those models. This is a Monte Carlo approach that treats theory-derived outlines as quite random--this is a principle of averaging across random paths of a variable in order to produce some expected value.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
no subject
Date: 2012-04-02 12:00 am (UTC)However, there are examples of mis-trust within the discipline.
I was surprised to learn that Bank of England uses several models with different theoretical foundations in order to forecast inflation and then provides the final estimate as a weighted average of outputs of those models. This is a Monte Carlo approach that treats theory-derived outlines as quite random--this is a principle of averaging across random paths of a variable in order to produce some expected value.