Sunday, April 24, 2016

The Distribution of Global Economic Activity...

... as proxied by the global distribution of nighttime lights (from a fascinating new paper by Hendersen et al.).  Like many good graphics, this one repays careful study.  You'll see lots of places where the lights match your prior, but you'll also see places that are perhaps "surprisingly" well-lit relative to popular perceptions (e.g., central America), other places that are perhaps surprisingly dark (e.g., most of Russia), fascinating patterns (e.g., look at Europe stretching east into Russia), etc.

Monday, April 18, 2016

On the Real-Time GDP War

A few days ago the WSJ did an interesting piece, Fed Banks Spar Over GDP Data, highlighting that the "race to provide credible real-time data on U.S. economic growth is pitting the Federal Reserve Bank of New York against its sibling in Atlanta."

In all this, real-time data on "economic growth" is interpreted as real-time data on GDP growth.

In my opinion, all of the real-time GDP products basically reflect a misguided perspective if the goal is real-time tracking of economic growth (which is as it should be, and what is claimed). If you want to track real-time growth, you should be tracking an extraction of a broad dynamic factor, effectively averaging over many indicators, not just tracking real-time GDP. That has been the leading and invaluable perspective from Burns and Mitchell straight through to modern dynamic-factor approaches.  My favorite, of course, is the FRB Philadelphia's ADS Index, but there are many others.

Wednesday, April 13, 2016

Big Data: Tall, Wide, and Dense

It strikes me that "tall", "wide", and "dense" might be useful words and conceptualizations of aspects of Big Data relevant in time-series econometrics.

Think of a  regression situation, with a  (T x K) "X matrix" for  T "days" (or whatever) of data for each of K variables.  Now imagine sampling intra-day, m times per day.  Then  X is (mT x K).  Big data correspond to huge-X situations arising because one or more of T, K, and m is huge, and they are usefully considered separately. (Of course there will always be subjectivity associated with "how huge is huge".)

-- As T gets large we have "tall data" (in reference to the tall X matrix, due to the large number of time periods, i.e., the long calendar span of data)

-- As K gets large we have "wide data" (in reference to the wide X matrix due to the large number of regressors)

-- As m gets large we have "dense data" (in reference to the high-frequency intra-day sampling, regardless of whether the data are tall)

A few examples:

--  Consider 2500 days of 1-minute returns for each of 5000 stocks.  The data are tall, wide and dense.

--  Consider 25 days of 1-minute returns for each of 50 stocks.  The data are dense, but neither tall nor wide.

--  Consider 2500 days of daily returns for each of 5000 stocks.   The data are tall and wide, but not dense.

Sunday, April 10, 2016

On "The Human Capital Approach to Inference"

Check out the interesting new paper by Bentley MacLeod at Columbia ("The Human Capital Approach to Inference"), on using economic theory in combination with machine learning to estimate conditional average treatment effects better than can be done with randomized control trials.

Quite apart from new methods for accurate estimation of conditional average treatment effects, the paper's intro contains some interesting tidbits on causal econometric inference. Here's one sequence in yellow, with my reactions:

BM: "There are two distinct approaches to modern empirical economics."
-- The MacLeod paper is exclusively about causal inference, so it should say "two distinct approaches to causal inference in modern empirical economics." Equating causal inference to all of empirical economics is simply wrong. Causal inference is a large and very important part of modern empirical economics, but far from its entirety. The booming field of financial econometrics, for example, is largely and intentionally reduced-form. See this.

BM: "First, there is research using structural models that begins by assuming individuals make utility maximizing decisions within a well defined environment, and then proceeds to measure the value of the unknown parameters..."
-- There is some unsettling truth here. A cynical but not-entirely-false view is that structural causal inference effectively assumes a causal mechanism, known up to a vector of parameters that can be estimated. Big assumption. And of course different structural modelers can make different assumptions and get different results.

BM: "The second approach addresses the self-selection of individuals into different observed treatments or choices by either explicitly randomizing treatments/choices in the context of an experiment...or through the use of a natural experiment that allows for an instrumental variables strategy. There is general agreement that explicit randomization provides one of the cleanest ways to obtain a measure of the effect of choice."
-- There's rarely general agreement about anything in economics. But yes, randomization is arguably the gold standard for causal effect estimation, if and when it can be done credibly.

Wednesday, March 30, 2016

Honoring Gregory Chow, Econometrics Pioneer

Wonderful to see this.  So many massive contributions. Structural change, optimal control, development, and much more.

New Issue for Frontiers of Economics in China (FEC)
Frontiers of Economics in China <>
Diebold, Francis X;
Tue 3/29/2016 9:02 PM
To help protect your privacy, some content in this message has been blocked. To re-enable the blocked features, click here.
To always show content from this sender, click here.

Full Text Available

Frontiers of Economics in China (FEC)

Dwight H. Perkins
Front. Econ. China. 2016, 11 (1): 1-6.   DOI: 10.3868/s060-005-016-0001-5
Abstract   PDF (71KB)

Zhiqi Chen
Front. Econ. China. 2016, 11 (1): 7-8.   DOI: 10.3868/s060-005-016-0002-2
Abstract   PDF (56KB)

Jushan Bai,Xu Han
Front. Econ. China. 2016, 11 (1): 9-39.   DOI: 10.3868/s060-005-016-0003-9
Abstract   PDF (311KB)

Yanqin Fan,Ruixuan Liu,Dongming Zhu
Front. Econ. China. 2016, 11 (1): 40-59.   DOI: 10.3868/s060-005-016-0004-6
Abstract   PDF (325KB)

Ming-Jen Chang,Ping Wang,Danyang Xie
Front. Econ. China. 2016, 11 (1): 60-87.   DOI: 10.3868/s060-005-016-0005-3
Abstract   PDF (3137KB)

Zili Yang
Front. Econ. China. 2016, 11 (1): 88-103.   DOI: 10.3868/s060-005-016-0006-0
Abstract   PDF (461KB)

Yongsheng Xu
Front. Econ. China. 2016, 11 (1): 104-122.   DOI: 10.3868/s060-005-016-0007-7
Abstract   PDF (375KB)

Ming Lei,Zihan Yin
Front. Econ. China. 2016, 11 (1): 123-141.   DOI: 10.3868/s060-005-016-0008-4
Abstract   PDF (186KB)

X. Henry Wang,Bill Z. Yang
Front. Econ. China. 2016, 11 (1): 142-155.   DOI: 10.3868/s060-005-016-0009-1
Abstract   PDF (327KB)

Yanrui Wu
Front. Econ. China. 2016, 11 (1): 156-172.   DOI: 10.3868/s060-005-016-0010-5
Abstract   PDF (283KB)

Frank H. Liu
Frontiers of Economics in China (FEC), Higher Education Press
F15, Fortune Tower 1, 4 Huixin East St., Chaoyang District, Beijing 100029, P. R. China
Tel:  010-58556312, 13911937357
QQ: 3542958,  Wechat: FEC2006

Tuesday, March 29, 2016

The Amazing Rise of Economic Measurement

Economic measurement, empirical economics, evidence-based economics -- call it what you want, but lately it's been trouncing the competition.  The table below, which speaks for itself, is from Bentley MacLeod's interesting new paper, "The Human Capital Approach to Inference" (more on it in a later post).  One can only wonder how its 2023 row will look.   (What a tectonic shift from the 70's and 80's.  I was a grad student in the 80's, when all the big profs were theorists and all the big students were aspiring theorists...)

Monday, March 28, 2016

Central Bank Forecast Accuracy

[Sorry for being AWOL.  Like everyone else, I'm generally three feet underwater and breathing through a straw, but in March and April it seems that even the straw goes under.  Anyway, lots of stuff in the pipeline, so let's try to get going again.]

Rummaging around in the basement I just found this Bloomberg piece on comparative central bank forecasting performance.  A friend emailed it a little more than six months ago (ouch), but it's still interesting.  Maybe you missed it.

Bloomberg's first-ever ranking of central bank forecasting, which is relied on by business and finance, turns up winners and losersBloomberg's first-ever ranking of central bank forecasting, which is relied on by business and finance, turns up winners and losers

Friday, March 11, 2016

Miserable Teaching Evaluations

I have always disliked teaching evaluations, feeling that they fail to measure true teaching effectiveness. And it's not just sour grapes -- really, I swear, I generally do fine and have won several teaching awards. Rather, I simply think that teaching evaluations create bad incentives. Ask yourself: Is the behavior that maximizes teaching evaluations the same behavior that maximizes true teaching effectiveness? No way.

But it may be much worse than that. Check out the abstract below for a seminar to be presented in Penn Statistics next week by Philip Stark, a 
Berkeley statistician (and Associate Dean of the Division of Mathematical and Physical Sciences). Paper here.


Teaching Evaluations (Mostly) Do Not Measure Teaching Effectiveness


Joint work with Anne Boring (SciencesPo) and Kellie Ottoboni (UC Berkeley)
Student evaluations of teaching (SET) are widely used in academic personnel decisions as a measure of teaching effectiveness. We show:
·         SET are biased against female instructors by an amount that is large and statistically significant
·         the bias affects how students rate even putatively objective aspects of teaching, such as how promptly assignments are graded
·         the bias varies by discipline and by student gender, among other things
·         it is not possible to adjust for the bias, because it depends on so many factors
·         SET are more sensitive to students' gender bias and grade expectations than they are to teaching effectiveness
·         gender biases can be large enough to cause more effective instructors to get lower SET than less effective instructors.
These findings are based on permutation tests applied to two datasets: 23,001 SET of 379 instructors by 4,423 students in six mandatory first-year courses in a five-year natural experiment at a French university, and 43 SET for four sections of an online course in a randomized, controlled, blind experiment at a US university. 

Wednesday, March 2, 2016

Georgetown Center for Econometric Practice

Check out the Georgetown Center for Econometric Practice (GCEP).  Web page here.  Facebook page here.  Screenshot below.



Georgetown Center for Econometric Practice
Tel: +1 202 687 6172



RiverviewOur training courses are designed to be of particular benefit to economists and social scientists in the public and private sectors wanting to know how to use econometric methods and a variety of data to inform policy making.
They run over 1 or 2 days in the new facilities of Georgetown University School of Continuing Studies located in downtown Washington, D.C. at 640 Massachusets Avenue N.W.