Halbert White

From InterSciWiki
Jump to: navigation, search
Halbert White.jpg

UCSD home

Jerry A. Hausman

John and Jennie S. MacDonald Professor, Department of Economics, Massachusetts Institute of Technology, Cambridge, MA USA WHY: with Halbert L. White, Jr., for their contributions to econometrics, specifically the Hausman specification test and the White standard errors test

Halbert L. White, Jr.

Chancellor's Associates Distinguished Professor of Economics, Department of Economics, University of California San Diego, La Jolla, CA USA WHY: with Jerry A. Hausman, for their contributions to econometrics, specifically the Hausman specification test and the White standard errors test

Can Mutual Fund “Stars” Really Pick Stocks? New Evidence from a Bootstrap Analysis R. Kosowski, A. Timmermann, H. White, and R. Wermers: Journal of Finance 61(6):2551 - 2595. 2006.

Compare this with Doyne Farmer Farmer, J D, P Patelli, and I I Zovko. "The Predictive Power of Zero Intelligence in Financial Markets" PNAS USA 102(11) (2005): 2254-2259.

Contents

The passing of a genius - March 31 2012

Dear Doug:

I just received an email about Hal passing away. I write this offering my condolences. I am so terribly sorry to hear about him passing and the way he passed. I hope it was better than I imagine it. I have been out of sorts the entire day. Hal had a great effect on my life as my undergraduate adviser. Working with him showed me the inner workings of the mind of a genius and how things can be so simple by thinking in a clear and organized way. His thoughts were clear and organized even though his office was busting at the seams with books that covered the walls and papers that covered the floor and desk. Sitting on his desk chair surrounded by papers taking notes while he stood near the tiny black board explaining modern portfolio theory. Trying to learn as much as I can before the phone rings again. The third or fourth door down on the second floor at the econ department in UCSD was his office. His small office was overflowing with his genius. Genius shined through his eyes and despite his humble smile, he humbled my then still arrogant teenage self. Other students saw economics through the classroom where he derived the demand function on four boards that were at the end of the lecture filled with Hal’s meticulous derivations, but I had the privilege of his guidance. He encouraged me to pursue higher mathematics and even though I did not end up taking functional analysis, I did declare a major in mathematics. Mathematics that helped me succeed in being an Economist and applied Econometrician. Econometrics I learned from him in his 200D course where he taught asymptotics. That course was only offered once a year and I had to take it while I was working and trying to graduate. With that course load I had little sleep and would have to go to work after so I was in my blazer in class at 8 AM. One day I woke myself up snoring in his class and I a line of drewel all the way on the lapel of my jacket and I woke up to see Hal looking at me and smiling. He was so kind. I hope he rests in peace. Please pass my condolences to his family. I would like to send a telegram, but I do not know who is the next of kin. Could you please help me.

Amer Aladhadh (apr3/12)

Halbert White 2012

Posted by James Hamilton at April 1, 2012 06:41 AM

Bates White Obituary

Bios - About Bates White

The Street

White’s life and accomplishments by viewing a video and article published by SignOnSanDiego (the website of the San Diego Union-Tribune).

From: (Video produced by Creative Services and Publications, UC San Diego.) 2011/oct 9

Economist Hal White plays trumpet, but he doesn’t blow his own horn. Which is to say, he doesn’t do a lot of bragging, even though there’s reason to.

Turns out that's a wise way to think. Lots of people -- notably those of Thomson Reuters -- thought he would win this year's Nobel Prize in economic sciences. But at 4 a.m. today, the prize was awarded to Thomas Sargent of NYU and Chris Sims of Princeton.


Halbert White, Jr. is an accomplished trumpet player in addition to being a world-class economist. Courtesy of the White family The Nobels are notoriously hard to predict, so its no surprise that this year's predictions about White were wrong. But TR notes that its prediction means that White could win this year, or in the near future. And White is still worthy of mention.

He’s had a direct effect on the life of everyone in this country.

In 1980, White shook economics by publishing a paper that lays out fundamental ways to evaluate the soundness of hypotheses posed by economists. It led to the White Standard Errors Test, which is in widespread use. The essence of the test is contained in that paper, which has been cited in more than 6,000 other papers, making it among the most influential papers in the history of modern economics.

Textbook writers struggle to explain White’s errors test in simple terms. White does not. He put it this way an an email:

“Economists apply mathematical models to real-world data to make predictions and to try to understand economic cause and effect. For example, a prediction answers the question, ‘What will the unemployment rate be next month if things continue the way they have been?’ Understanding cause-and-effect makes it possible to answer a question like, ‘What would happen to the unemployment rate if the Fed raises interest rates?’

“When economists build their models, they always build in a variety of assumptions. If these assumptions are wrong -- that is, they don’t match the data -- then the economists’ predictions won’t be as reliable as they might be, and answers to cause-and-effect questions may also be wrong. Sometimes this unreliability doesn’t matter much, and sometimes it can result in spectacular mistakes ...

“The way the test works is to compare two different estimates of ‘standard errors’. (These are numbers that measure the uncertainty associated with certain features of a model when it is applied to real-world data.) Although these estimates will always be different, they will be close to each other when the built-in assumptions are basically right, but they will be far apart when there is a problem. The White test checks whether the difference between these estimates is big enough to indicate that there is a problem. When there is, the exact nature of the problem can often be diagnosed by examining the pattern of differences in the standard error estimates.”

In other words, White prevents a lot of economists from goofing up.

The impact of this test, and other work by White, led Thomson Reuters to shortlist him for the Nobel Prize.

“White’s (1980) paper has been either the most cited or second most cited paper in economics over the last 30 years. That is an indicator of its great utility and influence,” said David Pendlebury, a citation analyst at Thomson Reuters.

“White has not published prolifically, but he’s published papers with a high average number of citations. He’s not just writing articles to write articles. White focuses on quality, not quantity.”

There’s another reason to think that White might win.

“The Nobel committee doesn’t like to give the prize in the same sub-field (every) year after year, and it has been awhile since they gave a Nobel for work in econometrics,” Pendlebury said. “It hasn’t happened since 2003, when Clive Granger and Rob Engle won the prize.”

There’s a couple of names you might recognize. At the time, Granger was a professor at UCSD, and Engle had just retired from the same faculty. They got “the call.” Today, the same may happen to Halbert White, Jr. If it does, the glory that reflects on UCSD will be due, in part, to different sort of call that Engle made to White in 1977.

In his bio, White says he was working at the University of Rochester when Engle called and “inquired if I might be interested in being a visiting assistant professor at UC San Diego. I had to think about that for a while -- perhaps 10 seconds ...

“In December of 1978 I flew out of Rochester during a blizzard and arrived in 75 degree sunshine to begin a visiting appointment.”

And, of course, to do a like trumpet playing.

Heteroskedastic Breusch-Pagan test

Heteroskedastic Breusch-Pagan test. H0: residuals homoskedastic

with Xun Lu: the Hausman test

PEARL, Greenland, Hal White, Xun Lu and others

Lots of papers at hand and at Judea Pearl, Sander Greenland and James M. Robins
  • H.White, Lu and Chalak
Lots of papers at hand and at Halbert White, Karim Chalak and Xun Lu
White, Halbert, and Xun Lu. 2010. "Robustness Checks and Robustness Tests in Applied Economics". Journal of Financial Econometrics, 8: 193-243. Matlab code
White and Lu 2009. Non-robustness is signaled by failure of conditional exogeneity. See Wooldridge 2006: 532-534, also http://www.oga-lab.net/RGM2/func.php?rd_id=systemfit:hausman.systemfit which might not be relevant. Lets look more into Hausman.
and more to come

On Causal modeling

Causal Discourse in a Game of Incomplete Information

Halbert White, Haiqing Xu, Karim Chalak. 2011. [Causal_Discourse_in_a_Game_of_Incomplete_Information.pdf Causal Discourse in a Game of Incomplete Information]. In progress. XuHaiqingECO20199.pdf

2010 Causality, Conditional Independence

Karim Chalak. Halbert White. 2010. Causality, Conditional Independence, and Graphical Separation in Settable Systems

Pearl Review of Causal graphs

Pearl, Judea 2010 The Foundations of Causal Inference: A Review (PDF) The Foundations of Causal Inference: A Review ∗ Judea Pearl † University of California, Los Angeles Computer Science Department Los Angeles, CA, 90095-1596, USA

2SLS and NSF Causality project

NSF Proposal 2010

Granger Causality: predictability in time series

H. White and X. Lu, "Granger Causality and Dynamic Structural Systems," submitted to Journal of Financial Econometrics memorial issue in honor of Clive Granger. .

White's Heteroscedasticity-Consistent Covariance Matrix Estimator

White, Halbert. 1980. A Heteroscedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroscedasticity. Econometrica 48:817-838. See: http://support.sas.com/rnd/app/examples/ets/hetero/index.htm

Data dredging

Boston college working papers with Chalak

  • White, Halbert, and Karim Chalak. "Identifying Effects of Endogenous Causes in Nonspearable Systems Using Covariates," submitted to Journal of Econometrics.
  • White, Halbert, and Karim Chalak. 2010. Testing a Conditional Form of Exogeneity,” Economics Letters, 109, 88-90.

Links

Retrieved from "http://intersci.ss.uci.edu/wiki/index.php?title=Halbert_White&oldid=74890"
Personal tools