Research
“Volatility Tests and Efficient Markets: A Review Essay”
Journal of Monetary Economics 27 (May 1991) 463-485. A review essay supposedly about Shiller’s book. It got me to think hard about volatility tests, and prove that they are exactly equivalent to regressions that forecast returns from price-dividend ratios.
Journal of Monetary Economics 27 (May 1991) 463-485. A review essay supposedly about Shiller’s book. It got me to think hard about volatility tests, and prove that they are exactly equivalent to regressions that forecast returns from price-dividend ratios.
A Critique of The Application of Unit Root Tests
Journal of Economic Dynamics and Control 15 (April 1991) 275-284. Running a battery of unit root/cointegration tests and then imposing the answers on subsequent analysis is a bad idea. Alas, there is no substitute for plotting the data and thinking about what makes sense.
Journal of Economic Dynamics and Control 15 (April 1991) 275-284. Running a battery of unit root/cointegration tests and then imposing the answers on subsequent analysis is a bad idea. Alas, there is no substitute for plotting the data and thinking about what makes sense.
What Should Macroeconomists Know About Unit Roots?
1991 NBER Macroeconomics Annual 6, (1991), 201-210. Comments on ‘Pitfalls and Opportunities: What Macroeconomists Should Know About Unit Roots’” by John Campbell, JSTOR Another blistering critique of the (mis) use of unit root econometrics.
1991 NBER Macroeconomics Annual 6, (1991), 201-210. Comments on ‘Pitfalls and Opportunities: What Macroeconomists Should Know About Unit Roots’” by John Campbell, JSTOR Another blistering critique of the (mis) use of unit root econometrics.
Production-Based Asset Pricing and the Link Between Stock Returns and Economic Fluctuations.
Journal of Finance 46 (1) (March 1991) 209-237. The q theory works pretty well if you difference it – investment growth is nicely correlated with stock returns, and the I/K ratio forecasts future stock returns. JSTOR
Journal of Finance 46 (1) (March 1991) 209-237. The q theory works pretty well if you difference it – investment growth is nicely correlated with stock returns, and the I/K ratio forecasts future stock returns. JSTOR
The Sensitivity of Tests of the Intertemporal Allocation of Consumption to Near-Rational Alternatives
American Economic Review 79 (June 1989) 319-337. Many tests of the permanent income model or consumption based asset pricing models exploit predictions that imply trivial utility costs. For example, adjusting consumption when you get the check rather than when you get the news can have utility costs of a few cents. Since our models abstract from small real-world costs and frictions, I proposed the idea of using the region of trivial utility costs as a measure of “economic standard errors” for model predictions.
American Economic Review 79 (June 1989) 319-337. Many tests of the permanent income model or consumption based asset pricing models exploit predictions that imply trivial utility costs. For example, adjusting consumption when you get the check rather than when you get the news can have utility costs of a few cents. Since our models abstract from small real-world costs and frictions, I proposed the idea of using the region of trivial utility costs as a measure of “economic standard errors” for model predictions.
The Return of the Liquidity Effect: A Study of the Short Run Relation Between Money Growth and Interest Rates
Journal of Business and Economic Statistics 7 (January 1989) 75-83. In the short run, we expect money growth and interest rates to be negatively correlated – the “liquidity effect.” In the long run, they should be positively correlated – the “inflation effect.” I used bandpass filters to isolate the “runs” and confirmed this prediction. This paper was part of my PhD thesis, and inspired by reading a misleading graph in a Wall Street Journal Op Ed that claimed we were in a new “super-neutrality” regime in which the correlation was always positive.
Journal of Business and Economic Statistics 7 (January 1989) 75-83. In the short run, we expect money growth and interest rates to be negatively correlated – the “liquidity effect.” In the long run, they should be positively correlated – the “inflation effect.” I used bandpass filters to isolate the “runs” and confirmed this prediction. This paper was part of my PhD thesis, and inspired by reading a misleading graph in a Wall Street Journal Op Ed that claimed we were in a new “super-neutrality” regime in which the correlation was always positive.
This was my job market paper, many years ago. Morals: write your thesis on something interesting, not a complex extension of your adviser's latest theory. Do something fun in your job market talk, like demonstrate bandpass filters by swinging your keys. Have Lars Hansen discover a really interesting mistake in technical Appendix C of your job market paper. You too might get a job at Chicago
Production Based Asset Pricing
1988. NBER working paper 2776. This one uses two technologies and two states to infer contingent claims prices from production decisions, and matches the equity premium and term premium. It has nothing to do with the “Production-based” papers that came later in the Journal of Finance and JPE. I abandoned the project because it’s too easy – there are no probabilities in firm decisions with this standard technology, so it’s very easy to get contingent claims prices that differ from probabilities.
Recent work by Frederico Belo and Urbann Jermann may finally break through the identification problems and make the approaches of these last two papers work.
1988. NBER working paper 2776. This one uses two technologies and two states to infer contingent claims prices from production decisions, and matches the equity premium and term premium. It has nothing to do with the “Production-based” papers that came later in the Journal of Finance and JPE. I abandoned the project because it’s too easy – there are no probabilities in firm decisions with this standard technology, so it’s very easy to get contingent claims prices that differ from probabilities.
Recent work by Frederico Belo and Urbann Jermann may finally break through the identification problems and make the approaches of these last two papers work.
How Big is the Random Walk in GNP?
Journal of Political Economy 96 (October 1988) 893-920. Short-order ARMA models suggest that GNP looks a lot like a random walk. But short-order ARMA models are fit to match one-step ahead forecasts, and can do a poor job of capturing long-term forecastability. I used a variance-ratio statistic (variance of long-term differences / variance of one-year differences) to show that there is a lot of mean-reversion in GNP that short-order ARMA models miss. I think the subsequent “permanent and transitory components” answers the substantive question better, but the warning about using long-term implications of short-term models remains worthwhile today.
Journal of Political Economy 96 (October 1988) 893-920. Short-order ARMA models suggest that GNP looks a lot like a random walk. But short-order ARMA models are fit to match one-step ahead forecasts, and can do a poor job of capturing long-term forecastability. I used a variance-ratio statistic (variance of long-term differences / variance of one-year differences) to show that there is a lot of mean-reversion in GNP that short-order ARMA models miss. I think the subsequent “permanent and transitory components” answers the substantive question better, but the warning about using long-term implications of short-term models remains worthwhile today.
Multivariate Estimates of the Permanent Components in GNP and Stock Prices
Journal of Economic Dynamics and Control, 12 (June/July 1988) 255-296. (With Argia M. Sbordone). This paper sits halfway between the “random walk in GNP” JPE and “permanent and transitory components” QJE. The “random walk” is univariate. Here, we realized that consumption could tell you a lot about the permanent component of GNP. Here, we use that insight in spectral and variance-ratio calculations. The answers are the same as in “permanent and transitory components”, but I now prefer the simpler VAR treatment in that paper. When GNP or stock prices are cointegrated with a random walk the subtle long-horizon and “nonparametric” techniques needed in the “random walk in GNP” really are no longer needed; short order models to produce good long-term forecasts.
Journal of Economic Dynamics and Control, 12 (June/July 1988) 255-296. (With Argia M. Sbordone). This paper sits halfway between the “random walk in GNP” JPE and “permanent and transitory components” QJE. The “random walk” is univariate. Here, we realized that consumption could tell you a lot about the permanent component of GNP. Here, we use that insight in spectral and variance-ratio calculations. The answers are the same as in “permanent and transitory components”, but I now prefer the simpler VAR treatment in that paper. When GNP or stock prices are cointegrated with a random walk the subtle long-horizon and “nonparametric” techniques needed in the “random walk in GNP” really are no longer needed; short order models to produce good long-term forecasts.