Thursday, January 21, 2021

Mixtures of Predictive Densities

Here's a new one, with Minchul Shin and Boyuan Zhang, "On the Aggregation of Probability Assessments: Regularized Mixtures of Predictive Densities for Eurozone Inflation and Real Interest Rates".

We propose methods for constructing regularized mixtures of density forecasts, exploring a variety of objectives and regularization penalties, and using them in a substantive exploration of Eurozone inflation and real interest rate density forecasts. The coolest thing is seeing what the regularization actually does. It has very different effects before and after the great recession. From the Great Recession onward, regularization moves density forecast probability mass from the centers to the tails, correcting for overconfidence. And it does so in real time, with no look-ahead cheating...

Tuesday, January 19, 2021

Welcome to 2021

 Welcome back!

I insist that 2021 will be better.  Here's a light one to start it off.  

I recently had occasion to visit the web site of Marcin Zamojski, a top young econometrician at the University of Gothenburg. He describes himself there as:

[I am a] frequentist at heart. I am also a strong believer in 'all models are wrong, but some are useful'. I am warming up to machine learning. If the data has a time dimension, count me in.

Wow!  Is that Marcin Zamoiski or Frank Diebold?

Wednesday, December 23, 2020

Until Next Year

File:Happy Holidays (5318408861).jpg

Your blogger will be back in the New Year. 

Meanwhile, Happy Holidays to all!

Best Webinar Awards, V (Newcomers and Recap)

Two more exciting econometrics webinars have recently burst on the scene:

Climate Econometrics:  Just what it sounds like -- the interface of climate science and econometrics

AMLEDS: "Applied Machine Learning, Economics, and Data Science".  So far mostly the interface of machine learning and econometrics.

Too soon to make any awards, but stay tuned for next year!


To recap, the last few posts have featured, in no particular order:

Chamberlain Seminar

Society for Financial Econometrics online seminar

International Association for Applied Econometrics webinar

FRBSF Virtual Seminar on Climate Economics

Climate Econometrics


What have I missed?

Best Webinar Awards, IV (SoFiE)

Now let's do the wonderful Society for Financial Econometrics online seminar.  It's a tie!  

The first winner is 

Rob Engle (NYU Stern), 


"Measuring and Hedging Geopolitical Risk,"

with S. Martins.  A fascinating "geopolitical risk" interpretation of the factor structure in "idiosyncratic" components of asset return volatilities.

Abstract:  Geopolitical events can impact volatilities of all assets, asset classes, sectors and countries. It is shown that innovations to volatilities are correlated across assets and therefore can be used to measure and hedge geopolitical risk. We introduce a definition of geopolitical risk which is based on volatility shocks to a wide range of financial market prices. To measure geopolitical risk, we propose a statistical model for the magnitude of the common volatility shocks. Accordingly, a test and estimation methods are developed and studied using both empirical and simulated data. We provide a novel explanation for why idiosyncratic volatilities comove based on a new way to formulate multiplicative factors. Finally, we propose a new criterion for portfolio optimality which is intended to reduce the exposure to geopolitical risk.

The second winner is 

Patrick Gagliardini (U Lugano), 


"Extracting Statistical Factors When Betas Are Time-Varying,"

with H. Ma.  Time-varying betas are a key route to time-varying financial network connectedness, near and dear to my heart.

Check out the videos here. 

This paper deals with identification and inference on the unobservable conditional factor space and its dimension in large unbalanced panels of asset returns. The model specification is nonparametric regarding the way the loadings vary in time as functions of common shocks and individual characteristics. The number of active factors can also be time-varying as an effect of the changing macroeconomic environment. The method deploys Instrumental Variables (IV) which have full-rank covariation with the factor betas in the cross-section. It allows for a large dimension of the vector generating the conditioning information by machine learning techniques. In an empirical application, we infer the conditional factor space in the panel of monthly returns of individual stocks in the CRSP dataset between January 1971 and December 2017.

Monday, December 21, 2020

Climate Finance

Very nice empirically-oriented (and empirically-sophisticated) Giglio-Kelly-Stroebel survey of climate change and financial markets here.  Ungated copy here.   

Sunday, December 20, 2020

Best Webinar Awards, III (IAAE)

Now let's do the always-stimulating International Association for Applied Econometrics webinar. The winner is:

Andrii Babii (UNC Chapel Hill), 


"Binary Choice with Asymmetric Loss and Fairness in
Machine Learning Classification, with an Application to Racial Justice,"

with Chen, Ghysels, and Kumar. Check out the paper here, and video+slides here.

Asymmetric loss is crucially relevant in some situations; consider, for example, classification as "guilty" or "non guilty". Traditional classification methods have a hard time with it, however, as they ultimately treat type I and II errors symmetrically. (See, e.g., here.) This paper makes impressive progress.


The importance of asymmetries in prediction problems arising in economics has been recognized for a long time. In this paper, we focus on binary choice problems in a data-rich environment with general loss functions. In contrast to the asymmetric regression problems, the binary choice with general loss functions and high-dimensional datasets is challenging and not well understood. Econometricians have studied binary choice problems for a long time, but the literature does not offer computationally attractive solutions in data-rich environments. In contrast, the machine learning literature has many computationally attractive algorithms that form the basis for much of the automated procedures that are implemented in practice, but it is focused on symmetric loss functions that are independent of individual characteristics. One of the main contributions of our paper is to show that the theoretically valid predictions of binary outcomes with arbitrary loss functions can be achieved via a very simple reweighting of the logistic regression, or other state-of-the-art machine learning techniques, such as boosting or (deep) neural networks. We apply our analysis to racial justice in pretrial detention.

See also here.

Friday, December 18, 2020

Best Webinar Awards, II (Chamberlain)

Now let's do the wonderful and pioneering Chamberlain Seminar.

The winner is:

Elena Manresa (NYU), 


"An Adversarial Approach to Structural Estimation,"

with Tetsuya Kaji and Guillaume Pouliot! Check out the paper and video+slides.

It knocked me off my feet (and a few others – there were 900+ viewers). The way I see it -- although the approach is actually much more sophisticated than the description I'm about to give -- she proposes and explores, theoretically and empirically, the use of machine learning (ML) approximators like neural nets (NNs), random forests, etc. as windows for indirect inference in structural econometric models. This is a big deal, as ML approximators are potentially very sophisticated tools for characterizing model and data properties, thereby sharpening our ability to detect divergences between them. Of course her paper raises many questions as well, as does all good research, for example whether the numerous local optima associated with NNs will complicate the resulting indirect inference estimation. In any event the work is tremendously stimulating – a long and exciting way from casual GMM based on a few moments selected in ad hoc fashion, and a very nice bridge between the econometrics and data science / ML literatures. The paper was a real “eureka moment” for me. 

We propose a new simulation-based estimation method, adversarial estimation, for structural models. The estimator is formulated as the solution to a minimax problem between a generator (which generates synthetic observations using the structural model) and a discriminator (which classifies if an observation is synthetic). The discriminator maximizes the accuracy of its classification while the generator minimizes it. We show that, with a sufficiently rich discriminator, the adversarial estimator attains parametric efficiency under correct specification and the parametric rate under misspecification. We advocate the use of a neural network as a discriminator that can exploit adaptivity properties and attain fast rates of convergence. We apply our method to the elderly’s saving decision model and show that including gender and health profiles in the discriminator uncovers the bequest motive as an important source of saving across the wealth distribution, not only for the rich.

Thursday, December 17, 2020

2020 Best Webinar Awards, I (FRBSF)

I'm sure you've been anxiously awaiting my (first annual?) "Best of 2020" econometrics retrospective! Let's do "best new webinars".  I'll list my top webinars in this and forthcoming posts (in no particular order) and select a "best talk" from each.  Of course they're filled with great talks -- that's why they're my favorite webinars -- quite apart from my personal selection for best talk.

Let's start with the Federal Reserve Bank of San Francisco's rock-solid Virtual Seminar on Climate Economics

And the winner is:

Solomon Hsiang (Berkeley), 


"Valuing the Global Mortality Consequences of Climate Change"!

Congrats to Sol and his 16 coauthors (yes, 16!) for producing a truly breathtaking global empirical analysis, blending massive observational data and climate model simulations to help inform a pressing issue of global importance.  Check out the paper and video

ABSTRACT This paper develops the first globally comprehensive and empirically grounded estimates of mortality risk due to future temperature increases caused by climate change. Using 40 countries' subnational data, we estimate age-specific mortality-temperature relationships that enable both extrapolation to countries without data and projection into future years while accounting for adaptation. We uncover a U-shaped relationship where extreme cold and hot temperatures increase mortality rates, especially for the elderly, that is flattened by both higher incomes and adaptation to local climate (e.g., robust heating systems in cold climates and cooling systems in hot climates). Further, we develop a revealed preference approach to recover unobserved adaptation costs. We combine these components with 33 high-resolution climate simulations that together capture scientific uncertainty about the degree of future temperature change. Under a high emissions scenario, we estimate the mean increase in mortality risk is valued at roughly 3.2% of global GDP in 2100, with today's cold locations benefiting and damages being especially large in today's poor and/or hot locations. Finally, we estimate that the release of an additional ton of CO2 today will cause mean [interquartile range] damages of $36.6 [-$7.8, $73.0] under a high emissions scenario and $17.1 [-$24.7, $53.6] under a moderate scenario, using a 2% discount rate that is justified by US Treasury rates over the last two decades. Globally, these empirically grounded estimates substantially exceed the previous literature's estimates that lacked similar empirical grounding, suggesting that revision of the estimated economic damage from climate change is warranted.

Tuesday, December 15, 2020