Lars Peter Hansen, Nobel Laureate

This is a prize given to time series econometrics and how to deal with imperfect data and changing variances for variables being estimated.  Can you say “Generalized Method of Moments” (GMM)?  Hansen teaches at the economics department of the University of Chicago.

For years now journalists have asked me if Hansen might win, and if so, how they might explain his work to the general reading public.  Good luck with that one.

Here is Hansen on  Here is Hansen’s home page.  Here is Hansen on Wikipedia.  Hansen developed GMM in 1982, and here are some lecture notes on GMM:

Unlike maximum likelihood estimation (MLE), GMM does not require complete knowledge of the distribution of the data. Only specified moments derived from an underlying model are needed for GMM estimation. In some cases in which the distribution of the data is known, MLE can be computationally very burdensome whereas GMM can be computationally very easy. The log-normal stochastic volatility model is one example. In models for which there are more moment conditions than model parameters, GMM estimation provides a straightforward way to test the specification of the proposed model. This is an important feature that is unique to GMM estimation.

Here is a highly technical piece on how GMM is useful for testing macroeconomic propositions.  This is a slightly more intuitive treatment than most sources.

If you read this piece with Hodrick, you will see that Hansen’s work is instrumental for testing the advanced versions of the propositions of Fama and Shiller. In this critical sense, the three prizes are quite tightly unified.  And see this paper too with Singleton.  Here’s the most concrete sentence you are going to squeeze out of me on this one: if you want to do serious analysis of whether changing risk premia can help rationalize observed asset price movements, Hansen’s contributions will prove essential.


I still think GMM is one of the coolest empirical approaches in economics. We can estimate one equation of a macro model without assuming that the entire model is "true." The computational demands are minimal. It's also closely related to SMM, which allows for calibrating more complicated models. These are workhorse approaches in macro, so I think this is a very good choice for the prize.

How about a concrete example showing how applying his theories leads to a unexpected or otherwise unknowable result?

Say you have a gamma distribution. The method of moments makes it way easier to estimate the mean and variance of a distribution using samples than using Maximum Likelihood Distribution.

I'm a probabilist, that's as concrete as I get.

But that's an example using just the very old method of moments. Hansen made some real advancements generalizing it to other functions-- before him the method of moments had largely been superseded by maximum likelihood estimation and other methods for many uses, but his generalization made it not only viable but preferred for lots of cases.

How can you get to that "concrete" sentence if you can't provide a reasonably cogent explanation of the inference method ? Better to say that you don't understand this at all than to make this claim.

A well deserved Nobel.

One underdeveloped implication of GMM for macro modeling is that calibration of macro models is essentially GMM with a very particular (and often extreme) choice of a weighting matrix. Thus macro modelers can calculate the standard errors and perform inference both the calibrated estimates as well as model predictions. Unfortunately most macro-practicioners do not perform or report the results of these tests. For more see Hansen and Heckman (1996).

The Nobel Prize committee honored Lars Peter Hansen for his work in developing a statistical method for testing rational theories of asset price movements. The statistical method Hansen developed is Generalized Method of Moments (GMM). The fact that Hansen won the Nobel Prize for his “empirical analysis of asset prices” caught me off guard as I did not realize this was the original application of GMM.

GMM is used in the estimation of the New Keynesian Phillips Curve. The New Keynesian Phillips Curve includes expectations of future inflation as an idependent variable. Since inflation expectations cannot really be observed, GMM offers a way around this difficulty.

The New Keynesian Phillips Curve, which was developed in 1995, is integral to most DSGE models that central banks across the globe are increasingly dependent. Thus it’s hard to imagine modern central banking without Hansen’s contributions to econometrics. So for Hansen to have won the prize for his empirical analysis of asset prices strikes me as somewhat ironic.


It makes sense that the finance applications would be cited given that Hansen is being linked with Fama and Shiller. This looks like a prize driven by Per Krusell, apparently the successor to Jurgen Weibull as dominant figure of the economics Nobel committee. Krusell is sympathetic wtih the Minnesota DSGE approach, and Hansen's GMM is used indeed for the calibrations done by the harder core followers of that approach.

As it is, prior to 2008, year after year Fama topped the list for those predicing the Nobel, particularly in betting markets, but did not get it. After the 2008 crash, his star fell and that of behavioralist Shiller rose, with many talking of a Shiller-Thaler prize. I began commenting on blogs some time ago (I think including here a few times) that the committee might replicate the surprise of the old Myrdal-Hayek award by pairing Fama with Shiller, although I figured Thaler would be thrown in for good measure. As it is, the balancing act of the orthodox Fama and the behavioralist Shiller is completed with the empiricist Hansen, who tilts to the Fama side of things.

As it is, I do think Hansen worthy ol the prize for GMM, but most of those forecasting him had him either alone or with other macroeconometricians. This is an ingenious combination, showing both diplomacy in bowing to both sides of the asset pricing debate while throwing in the serious empricist, Hansen.

Thanks for those two posts. Hansen winning a Nobel is no surprise, nor Fama and Shiller, but I couldn't figure out what connected the three of them. It was easy enough to guess that the answer was some sort of connection between GMM and empirical finance, but I didn't know exactly what that connection was.

Comments for this post are closed