Abstract:Using a generalized cross-spectral approach, we propose a model-free omnibus statistical procedure to check whether the direction of changes in an economic variable is predictable using the history of its past changes. A class of separate inference procedures are also given to gauge possible sources of directional predictability. They can reveal information about whether the direction of future changes is predictable using the direction, level, volatility, skewness, and kurtosis of past changes. An important feature of the proposed procedures is that they check many lags simultaneously, which is particularly suitable for detecting the alternatives whose directional dependence is small at each lag but it carries over a long distributional lag. At the same time, the tests naturally discount higher order lags, which is consistent with the conventional wisdom that financial markets are more influenced by the recent past events than by the remote past events. We apply the proposed procedures to four daily U.S. stock price indices. We find overwhelming evidence that the directions of excess stock returns are predictable using past excess stock returns, and the evidence is stronger for the directional predictability of large excess stock returns. In particular, the direction and level of past excess stock returns can be used to predict the direction of future excess stock returns with any threshold, and the volatility, skewness and kurtosis of past excess stock returns can be used to predict the direction of future excess stock returns with nonzero thresholds (i.e., large returns). The well-known strong volatility clustering together with weak serial dependence in mean cannot completely explain all documented directional predictability for stock returns. To exploit the economic significance of the documented directional predictability for stock returns, we consider a class of autologit models for directional forecasts and find that they have significant out-of-sample directional predictive power. Some trading strategies based on these models and their combinations can earn significant out-of-sample extra risk-adjusted returns over the buy-and-hold trading strategy

Predictive Regressions: Method of Least Autocorrelation

Predictive Regressions: Method of Least Autocorrelation

Mildly Explosive Autoregression with Anti-persistent Errors

Improving Autocorrelation Regression for the Hurst Parameter Estimation of Long-Range Dependent Time Series Based on Golden Section Search

Predictability of Equity Returns over Different Time Horizons: A Nonparametric Approach

Consistent Testing for Serial Correlation of Unknown Form

With random regressors, least squares inference is robust to correlated errors with unknown correlation structure

Predictive regression with p -lags and order- q autoregressive predictors

Balanced Predictive Regressions

Good volatility, bad volatility, and time series return predictability

Bias correction based on AR model in spurious regression

On Robust Inference in Time Series Regression

Are the Directions of Stock Price Changes Predictable? A Generalized Cross-Spectral Approach

Electronic structure study by means of X-ray spectroscopy and theoretical calculations of the "ferric star" single molecule magnet

A Regularized High-Dimensional Positive Definite Covariance Estimator with High-Frequency Data

MORE EFFICIENT ESTIMATION IN NONPARAMETRIC REGRESSION WITH NONPARAMETRIC AUTOCORRELATED ERRORS

Least Absolute Relative Error Estimation

On LASSO for predictive regression

Estimating spatial autoregressions under heteroskedasticity without searching for instruments

A NAIVE LEAST SQUARES METHOD FOR SPATIAL AUTOREGRESSION WITH COVARIATES

Approximate Factor Models with Strongly Correlated Idiosyncratic Errors