Implementing Statistical Criteria to Select Return Forecasting Models: What Do We Learn?
- 1 April 1999
- journal article
- research article
- Published by Oxford University Press (OUP) in The Review of Financial Studies
- Vol. 12 (2), 405-428
- https://doi.org/10.1093/rfs/12.2.405
Abstract
Statistical model selection criteria provide an informed choice of the model with best external (i.e., out-of-sample) validity. Therefore they guard against overfitting (“data snooping”). We implement several model selection criteria in order to verify recent evidence of predictability in excess stock returns and to determine which variables are valuable predictors. We confirm the presence of in-sample predictability in an international stock market dataset, but discover that even the best prediction models have no out-of-sample forecasting power. The failure to detect out-of-sample predictability is not due to lack of power.Keywords
This publication has 18 references indexed in Scilit:
- Assessing Goodness-Of-Fit of Asset Pricing Models: The Distribution of the Maximal R 2The Journal of Finance, 1997
- The Risk and Predictability of International Equity ReturnsThe Review of Financial Studies, 1993
- Simple Technical Trading Rules and the Stochastic Properties of Stock ReturnsThe Journal of Finance, 1992
- Efficient Capital Markets: IIThe Journal of Finance, 1991
- Data-Snooping Biases in Tests of Financial Asset Pricing ModelsThe Review of Financial Studies, 1990
- Economic Significance of Predictable Variations in Stock Index ReturnsThe Journal of Finance, 1989
- THE ESTIMATION OF THE ORDER OF AN AUTOREGRESSION USING RECURSIVE RESIDUALS AND CROSS‐VALIDATIONJournal of Time Series Analysis, 1989
- Stock returns and the term structureJournal of Financial Economics, 1987
- Predicting returns in the stock and bond marketsJournal of Financial Economics, 1986
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974