Yue Fang

Forecasting combination and encompassing tests

Abstract In this paper we demonstrate that forecast encompassing tests are valuable tools in getting an insight into why competing forecasts may be combined to produce a composite forecast which is superior to the individual forecasts. We also argue that results from forecast encompassing tests are potentially useful in model specification. We illustrate this using forecasts of quarterly UK consumption expenditure data from three classes of models: ARIMA, DHSY and VAR models.

The predictability of asset returns: an approach combining technical analysis and time series forecasts

We investigate predictability of asset returns by developing an approach that combines technical analysis and conventional time series forecasts. While exploiting predictable components as functions of past prices or returns, technical trading rules and time series forecasts capture different aspects of market predictability: the former tends to identify periods to be in the market when returns are positive and the latter is capable of identifying periods to be out when returns are negative. Applied to daily Dow Jones Averages over the first 100 years, the combined strategies outperform both technical trading rules and time series forecasts. The predictability can be explained largely by non-trivial low-order serial correlations in returns and is not mainly attributable to measurement errors arising from non-synchronous trading.

Generalized least squares with misspecified serial correlation structures

Summary. The regression literature contains hundreds of studies on serially correlated disturbances. Most of these studies assume that the structure of the error covariance matrix Q is known or can be estimated consistently from data. Surprisingly, few studies investigate the properties of estimated generalized least squares (GLS) procedures when the structure of Q is incorrectly identified and the parameters are inefficiently estimated. We compare the finite sample efficiencies of ordinary least squares (OLS), GLS and incorrect GLS (IGLS) estimators. We also prove new theorems establishing theoretical efficiency bounds for IGLS relative to GLS and OLS. Results from an exhaustive simulation study are used to evaluate the finite sample performance and to demonstrate the robustness of IGLS estimates vis-a-vis OLS and GLS estimates constructed for models with known and estimated (but correctly identified) Q. Some of our conclusions for finite samples differ from established asymptotic results.

The compass rose and random walk tests

The recent discovery of the compass rose pattern (Crack and Ledoit J. Finance 51(2) (1996) 751) has sparked considerable interest among researchers. This paper explores the significance of the effect of the compass rose pattern on random walk tests and measures to what extent its influence may limit the performance of test statistics. We show that in general, the asymptotic theory of test statistics is invalid for transactions data. However, Monte Carlo simulations indicate that the impact of the pattern, measured by the empirical size, is visible for moderate size samples only when the tick/volatility ratio is above some threshold, a condition that is readily met with intraday but not daily or weekly returns.

Forecasting with serially correlated regression models

In this article we investigate the asymptotic and finite-sample properties of predictors of regression models with autocorrelated errors. We prove new theorems associated with the predictive efficiency of generalized least squares (GLS) and incorrectly structured GLS predictors. We also establish the form associated with their predictive mean squared errors as well as the magnitude of these errors relative to each other and to those generated from the ordinary least squares (OLS) predictor. A large simulation study is used to evaluate the finite-sample performance of forecasts generated from models using different corrections for the serial correlation.

GMM tests for the Katz family of distributions

Generalized method of moments (GMM) is used to develop tests for discriminating discrete distributions among the two-parameter family of Katz distributions. Relationships involving moments are exploited to obtain identifying and over-identifying restrictions. The asymptotic relative efficiencies of tests based on GMM are analyzed using the local power approach and the approximate Bahadur efficiency. The paper also gives results of Monte Carlo experiments designed to check the validity of the theoretical findings and to shed light on the small sample properties of the proposed tests. Extensions of the results to compound Poisson alternative hypotheses are discussed.

Block-crossed arrays with applications to robust design

Abstract In this paper we develop a class of two-level designs that allow orthogonal estimation of certain main effects and two-factor interactions when other main effects and two-factor interactions are present but not of interest. The designs can be easily generated using block-crossed arrays. The application is clearly to settings in which one might apply either Taguchi’s crossed arrays or single-array designs suggested by many people as a better alternative to Taguchi’s methods. Our designs have certain properties, the most important being the flexibility and a reduction, in many cases, in the number of runs over both Taguchi’s designs and single orthogonal arrays.

Using least squares to generate forecasts in regressions with serial correlation

The topic of serial correlation in regression models has attracted a great deal of research in the last 50 years. Most of these studies have assumed that the structure of the error covariance matrix omega was known or could be consistently estimated from the data. In this article, we describe a new procedure for generating forecasts for regression models with serial correlation based on ordinary least squares and on an approximate representation of the form of the autocorrelation. We prove that the predictors from this specification are asymtotically efficient under some regularity conditions. In addition, we show that there is not much to be gained in trying to identify the correct form of the serial correlation since efficient forecasts can be generated using autoregressive approximations of the autocorrelation. A large simulation study is also used to compare the finite sample predictive efficiencies of this new estimator vis-a-vis estimators based on ordinary least squares and generalized least squares. Copyright 2008 The Authors

Semi-parametric specification tests for mixing distributions

We present a semi-parametric method for testing mixing distributions in the mixed Poisson model. The proposed method, which is based on the generalized method of moments, does not demand the complete specification of the probability function but only requires a specification of a set of moment conditions which the model should satisfy. We demonstrate that an explicit expression for moment relations between the mixing and the mixed distributions provides a natural way in selecting moment restrictions and model parameterization. The Monte Carlo evidence suggests that the test has satisfactory performance for moderate size samples.

Stock returns: momentum, volatility and interest rates

Various theories have been proposed to explain momentum in stock returns. We provide evidence in favor of risk-based explanations. Specifically, we construct self-financing market neutral portfolios that take long positions in past winners and short positions in past losers. We show that the return spreads between past winners and losers in the first year are driven primarily by high volatility stocks. Momentum investment strategies, which buy past winning stocks and sell past losing stocks, are significantly less profitable once we control for volatility. We also show that momentum profits appear to be associated with economic cycles as proxied by the prime rate. The momentum strategies are substantially stronger during expansions than during recessions. This is mainly due to the relatively poor performance of past losers (rather than superior performance of the past winners) during expansions.