Forecasting Autoregressive Time Series in the Presence of Deterministic Components

This paper studies the error in forecasting an autoregressive process with a deterministic component. We show that when the data are strongly serially correlated, forecasts based on a model that detrends the data using OLS before estimating the autoregressive parameters are much less precise than those based on an autoregression that includes the deterministic components, and the asymptotic distribution of the forecast errors under the two-step procedure exhibits bimodality. We explore the conditions under which feasible GLS trend estimation can lead to forecast error reduction. The finite sample properties of OLS and feasible GLS forecasts are compared with forecasts based on unit root pretesting. The procedures are applied to 15 macroeconomic time series to obtain real time forecasts. Forecasts based on feasible GLS trend estimation tend to be more efficient than forecasts based on OLS trend estimation. A new finding is when a unit root pretest rejects non-stationarity, use of GLS yields smaller forecast errors than OLS. When the series to be forecasted is highly persistent, GLS trend estimation in conjunction with unit root pretests can lead to sharp reduction in forecast errors.

[1]  K. Mynbaev.  Asymptotic Properties of OLS Estimates in Autoregressions with Bounded or Slowly Growing Deterministic Trends , 2006 .

[2]  K. Hadri,et al.  Is More Information a Good Thing? Bias Nonmonotonicity in Stochastic Difference Equations , 2000 .

[3]  G. Elliott.  Efficient Tests for a Unit Root When the Initial Observation Is Drawn from Its Unconditional Distribution , 1999 .

[4]  Francis X. Diebold,et al.  Unit-Root Tests Are Useful for Selecting Forecasting Models , 1999 .

[5]  J. Stock,et al.  A Comparison of Linear and Nonlinear Univariate Models for Forecasting Macroeconomic Time Series , 1998 .

[6]  Timothy J. Vogelsang,et al.  Trend Function Hypothesis Testing in the Presence of Serial Correlation , 1998 .

[7]  Francis X. Diebold,et al.  Elements of Forecasting , 1997 .

[8]  K. Abadir,et al.  Two Mixed Normal Densities from Cointegration Analysis , 1997 .

[9]  Mark W. Watson,et al.  Estimating Deterministic Trends in the Presence of Serially Correlated Errors , 1994, Review of Economics and Statistics.

[10]  P. Phillips,et al.  COWLES FOUNDATION FOR RESEARCH IN ECONOMICS , 1998 .

[11]  K. West,et al.  Asymptotic Inference about Predictive Ability , 1996 .

[12]  Colin P. Hargreaves,et al.  Non-Stationary Time Series Analysis and Cointegration , 1994 .

[13]  James D. Hamilton.  Time Series Analysis , 1994 .

[14]  J. Stock,et al.  Efficient Tests for an Autoregressive Unit Root , 1992 .

[15]  D. Andrews.  Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation , 1991 .

[16]  Michael Sampson,et al.  The Effect of Parameter Uncertainty on Forecast Variances and Confidence Intervals for Unit Root and Trend Stationary Time-Series Models , 1991 .

[17]  P. J. Green,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[18]  Peter C. B. Phillips,et al.  The sampling distribution of forecasts from a first-order autoregression , 1979 .

[19]  D. Rubinfeld,et al.  Econometric models and economic forecasts , 2002 .

[20]  Michael D. Geurts,et al.  Time Series Analysis: Forecasting and Control , 1977 .

[21]  Zvi Griliches,et al.  Small-Sample Properties of Several Two-Stage Regression Methods in the Context of Auto-Correlated Errors , 1969 .

[22]  A. Goldberger.  Best Linear Unbiased Prediction in the Generalized Linear Regression Model , 1962 .

[23]  U. Grenander,et al.  Statistical analysis of stationary time series , 1958 .