论文信息 - How far can changes in general business activity be forecasted

How far can changes in general business activity be forecasted

Abstract An F test [Nelson (1976)] of Parzen's prediction variance horizon [Parzen (1982)] of an ARMA model yields the number of steps ahead that forecasts contain information (short memory). A special 10 year pattern in Finnish GDP is introduced as a ‘seasonal’ in an ARMA-model. Forecasts three years ahead are statistically informative but exploiting the complete 10 year pattern raises doubts both about model memory and model validity.

Lars-Erik Öller | Lars-Erik Öller

[1] R. E. Kalman,et al. A New Approach to Linear Filtering and Prediction Problems , 2002 .

[2] Lars-Erik Öller,et al. Macroeconomic forecasting with a vector arima model : A case study of the finnish economy , 1985 .

[3] G. C. Tiao,et al. Modeling Multiple Time Series with Applications , 1981 .

[4] J. Geweke,et al. THE ESTIMATION AND APPLICATION OF LONG MEMORY TIME SERIES MODELS , 1983 .

[5] C. Granger,et al. Forecasting Economic Time Series. , 1988 .

[6] Emanuel Parzen,et al. ARARMA models for time series analysis and forecasting , 1982 .

[7] Per Hokstad,et al. A METHOD FOR DIAGNOSTIC CHECKING OF TIME SERIES MODELS , 1983 .

[8] A. I. McLeod,et al. Preservation of the rescaled adjusted range: 1. A reassessment of the Hurst Phenomenon , 1978 .

[9] Charles R. Nelson,et al. The Interpretation of R 2 in Autoregressive-Moving Average Time Series Models , 1976 .

[10] C. Granger. Long memory relationships and the aggregation of dynamic models , 1980 .

[11] Gwilym M. Jenkins,et al. Time series analysis, forecasting and control , 1972 .

[12] Robert L. McLaughlin. A model of an average recession and recovery , 1982 .

[13] Andrew Harvey,et al. A unified view of statistical forecasting procedures , 1984 .