What makes some risk models better than others, and is there a "best" model?

Recently I received a comment on my February 4 entry "Really, how different are the risk model providers?" that I think is worth addressing.

The anonymous comment read:

You refer here to the concept of a "best" risk model. Can you expand a bit on this? What does this mean to you? Surely the determination of "best" is in the interpretation and use of the model, rather than the model's construction?

In my post, I am careful to say, “My task isn’t to suggest which model is 'best.' Frankly, my analysis doesn’t offer genuine insight on that question.” I specifically picked these words because I don’t wish to imply that I do address the concept of “best.”

I believe that model construction is essential to the concept of a “best” risk model. Even though the analysis is ex ante, I strongly disagree with the notion that it is therefore wrong or impossible to judge a model. Investment managers should gravitate towards models that are transparently documented and supported by ongoing research efforts (e.g., industry conferences, white papers) because it demonstrates a provider’s confidence in and commitment to the model construction. Maybe there is or isn’t a clear “best” model, but there are undoubtedly “better” models and “lesser” risk models available to the investment management community.

I do agree that interpretation and use of the model is extremely important too.

Thank you for your comments. Please share your feedback below.

To receive future posts by e-mail, subscribe to this blog.