## Predictive regressions (1999)

Venue: | Journal of Financial Economics |

Citations: | 326 - 14 self |

### BibTeX

@ARTICLE{Stambaugh99predictiveregressions,

author = {Robert F. Stambaugh},

title = {Predictive regressions},

journal = {Journal of Financial Economics},

year = {1999},

pages = {375--421}

}

### Years of Citing Articles

### OpenURL

### Abstract

When a rate of return is regressed on a lagged stochastic regressor, such as a dividend yield, the regression disturbance is correlated with the regressor's innovation. The OLS estimator's "nite-sample properties, derived here, can depart substantially from the standard regression setting. Bayesian posterior distributions for the regression parameters are obtained under speci"cations that di!er with respect to (i) prior beliefs about the autocorrelation of the regressor and (ii) whether the initial observation of the regressor is speci"ed as "xed or stochastic. The posteriors di!er across such speci"cations, and asset allocations in the presence of estimation risk exhibit sensitivity to those

### Citations

2737 |
Time series analysis
- Hamilton
- 1994
(Show Context)
Citation Context ...as a standard Bayesian multivariate regression model ifR.F. Stambaugh / Journal of Financial Economics 54 (1999) 375}421 407 the &pre-sample' observations are assumed to be deterministic (see, e.g., =-=Hamilton, 1994-=-, p. 358). When modi"ed for the case of N equations, the prior in (23) becomes p(b, Σ)J�Σ���������. (47) This Bayesian VAR speci"cation with multiple predictive variables is used in the analyses of as... |

2564 | Equation of state calculations by fast computing machines - Metropolis, Rosenbluth, et al. - 1953 |

2217 | Time Series Analysis: Forecasting and Control - Box, Jenkins - 1970 |

1429 |
Statistical Decision Theory and Bayesian Analysis
- Berger
- 1985
(Show Context)
Citation Context ...lues of E(β�D) are reported here for several alternative speci"cations of the prior and the likelihood. The posterior mean has minimum posterior expected loss under a squared-error loss function (see =-=Berger, 1985-=-). In addition, the posterior density yields probabilities for composite hypotheses, such as prob(β)0), and, as will be observed, the inferences associated with such probabilities can contrast with th... |

1425 |
An Introduction in Multivariate Statistical Analysis
- Anderson
- 1958
(Show Context)
Citation Context ...����)��� "�Ψ���� "�< � ����. (A.46)420 R.F. Stambaugh / Journal of Financial Economics 54 (1999) 375}421 The last equality follows from the formula for the determinant of a partitioned matrix (e.g., =-=Anderson, 1984-=-, Theorem A.3.2). The Jacobian of the transformation from Σ�� to Σ is �Σ�������, and multiplying (A.46) by that quantity gives (52). If (22) is applied separately for b and Σ, as discussed at the end ... |

1385 | Monte Carlo Sampling Methods Using Markov Chains and Their - Hastings - 1970 |

1194 |
Bayesian Theory
- Bernardo, Smith
- 2000
(Show Context)
Citation Context ...tion that, in the simplest i.i.d. normal univariate setting, the number of degrees of freedom in the posterior for the variance is una!ected by whether the mean is known or unknown (see, for example, =-=Bernardo and Smith, 1994-=-, Chapter 5). In a time-series setting, the dichotomy between location and scale parameters is blurred. For example, ρ a!ects the conditional mean as well as the unconditional variance of x � . Philli... |

556 | Dividend yields and expected stock returns - Fama, French - 1988 |

554 |
Bayesian Inference in Statistical Analysis
- Box, Tiao
- 1973
(Show Context)
Citation Context ...e intervals correspond to Bayesian highest-posterior-density regions under di!use priors, and the p-value for a positive one-tailed test of β"0 is identical to the posterior probability that β)0 (see =-=Box and Tiao, 1973-=-). That correspondence no longer obtains in the current setting, wherein a Bayesian could, for example, assign low probability to β)0 at the same time the frequentist accepts that hypothesis because i... |

502 | Understanding the Metropolis-Hastings algorithm - Chib, Greenberg - 1995 |

428 | Stock returns and the term structure - Campbell - 1987 |

359 | Investing for the Long Run When Returns are Predictable - Barberis - 1999 |

296 | Forward and spot exchange rates - Fama - 1984 |

271 | Exchange Rates and Fundamentals: Evidence on Long-Horizon Predictability - Mark - 1995 |

268 | Predicting returns in the stock and bond markets - Keim, Stambaugh - 1986 |

257 | On the Predictability of Stock Returns: An Asset Allocation Perspective - Kandel, Stambaugh - 1996 |

240 | The Information in Long Maturity Forward Rates - Fama, Bliss - 1987 |

134 | Predictable Stock Returns: The Role of Small Sample Bias - Nelson, Kim - 1993 |

112 | On biases in tests of the expectations hypothesis of the term structure of interest rates - Bekaert, Hodrick, et al. - 1997 |

107 | The Information in the Term Structure - Fama - 1984 |

107 | Book-to-market, dividend yield, and expected market returns: A time-series analysis - Kothari, Shanken - 1997 |

100 | Computing the distribution of quadratic forms in normal variables - Imhof - 1961 |

79 | Note on the Bias in the Estimation of Autocorrelation - Kendall - 1954 |

74 | Inference in Models with nearly Integrated Regressors”, Econometric Theory 11 - Cavanagh, Elliott, et al. - 1995 |

70 | Forward Rates and Future Policy: Interpreting the Term Structure of Interest Rates - Shiller, Campbell, et al. - 1983 |

68 | Book-to-market ratios as predictors of market returns - Ponti, Schall - 1998 |

65 | Understanding unit rooters: A helicopter tour, Econometrica, forthcoming - Uhlig - 1991 |

64 | To Criticize the Critics: An Objective Bayesian Analysis of Stochastic Trends - Phillips - 1991 |

56 | Bayesian Skepticism on Unit Root Econometrics - Sims - 1988 |

49 | Inference in time series regression when the order of integration of a regressor is unknown, Econometric Theory 10 - Elliott, Stock - 1994 |

45 | Bias in Regressions with Lagged Stochastic Regressors. Working Paper - Stambaugh - 1986 |

41 | Do we reject too often? small sample properties of tests of rational expectations models - Mankiw, Shapiro - 1986 |

32 | Bias in the estimation of autocorrelations - Marriott, Pope - 1954 |

24 | Some observations on inverse probability including a new indifference rule - PERKS - 1947 |

20 | Bias in the estimation of multivariate autoregressions - Nicholls, Pope |

15 | Asset Returns and Ination - Fama, Schwert - 1977 |

15 | The Exact Moment of the Least Square Estimator for the Autoregressive Model - Sawa - 1978 |

11 | The Exact Moments of a Ratio of Quadratic Forms in Normal Variables,” Annales D’économie et de Statistique - Magnus - 1986 |

10 | Dividend yields are equity risk premiums - Roze - 1984 |

6 | The ”speculative e…ciency” hypothesis - Bilson - 1981 |

6 | The exact moments of the least-squares estimator for the autoregressive model: Corrections and extensions - Nankervis, Savin - 1988 |

6 | On Jeffreys prior when using the exact likelihood function. Econometric Theory - Uhlig - 1994 |

3 | On formulae for con dence points based on integrals of weighted likelihoods - Welch, Peers - 1963 |

1 | Currency hedging over short and long horizons with time-varying expected returns. Working Paper - Bauer - 1998 |

1 | Comment on &To criticize the critics - Leamer - 1991 |

1 |
Bayesian approaches to the &unit root' problem: a comment
- Stock
- 1991
(Show Context)
Citation Context ...ys information about the data in an objective fashion, and one might argue that the dependence on prior beliefs makes Bayesian analysis less e!ective in communicating a description of the data (e.g., =-=Stock, 1991-=-). On the other hand, one might argue that reporting implications for decisions describes the data in a more relevant manner, and a Bayesian framework is better suited to that purpose. Kandel and Stam... |