论文信息 - Forecasting Chilean Industrial Production and Sales with Automated Procedures

Forecasting Chilean Industrial Production and Sales with Automated Procedures

This paper presents a rigurous framework for evaluating alternative forecasting methods for Chilean industrial production and sales. While nonlinear features appear to be important for forecasting the very short term, simple univariate linear models perform about as well for almost every forecasting horizon.

Rómulo A. Chumacero

[1] Greg Tkacz. Neural network forecasting of Canadian GDP growth , 2001 .

[2] Paul Newbold,et al. Testing the equality of prediction mean squared errors , 1997 .

[3] F. Diebold,et al. Comparing Predictive Accuracy , 1994, Business Cycles.

[4] Yue Fang,et al. Forecasting combination and encompassing tests , 2003 .

[5] H. White,et al. Reply to comments on “artificial neural networks: an econometric perspective“ , 1994 .

[6] Joseph P. Romano,et al. The stationary bootstrap , 1994 .

[7] Lutz Kilian,et al. On the Selection of Forecasting Models , 2003, SSRN Electronic Journal.

[8] L. Kilian,et al. In-Sample or Out-of-Sample Tests of Predictability: Which One Should We Use? , 2002, SSRN Electronic Journal.

[9] Andrew J. Patton,et al. Properties of Optimal Forecasts , 2003 .

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

[11] Michael P. Clements,et al. FORECASTING ECONOMIC TIME SERIES , 2000, Econometric Theory.

[12] B. Hansen,et al. Inference in TAR Models , 1997 .

[13] H. White,et al. A Reality Check for Data Snooping , 2000 .