This text on economic forecasting asks why some practices seem to work empirically despite a lack of formal support from theory. After reviewing the conventional approach to forecasting, it looks at the implications for causal modelling, presents forecast errors and delineates sources of failure.
Numerical distribution functions of likelihood ratio tests for cointegration. Journal of Applied Econometrics, 14, 563–577. McCabe, B., & Tremayne, A. R. (1993). Elements of modern asymptotic theory with statistical applications.
McCabe, B. and Tremayne, A.R. (1993) Elements of Modern Asymptotic Theory with Statistical Applications. Manchester: Manchester University Press. Mosconi, R. and Giannini, C. (1992) Non-causality in cointegrated systems: representation, ...
Nonstationary Time Series Analysis and Cointegration shows major developments in the econometric analysis of the long run (of nonstationarity and cointegration) - a field which has developed dramatically over the...
Forecasting is required in many situations.
Durland , J. M. and McCurdy , T. H. ( 1994 ) . Duration dependent transitions in a Markov model of US GNP growth . Journal of Business and Economic Statistics , 12 , 279–288 . Emerson , R. A. ( 1994 ) . Two essays on Investment Trusts ...
Introduction to the theory of time series; Spectral analysis; Building linear series models; The theory of forecasting; Practical methods for univariate time series forecasting; Forecasting from regression models; Multiple modeling...
The results of this section can be generalized, as discussed by Engel [1984], where conditions are provided to assure that the sum of two dependent Gaussian ARMA processes is ARMA. Engel further shows that if X, is Gaussian ARMA with ...
From the author of the bestselling "Analysis of Time Series," Time-Series Forecasting offers a comprehensive, up-to-date review of forecasting methods.
... methods of analyzing a time series; both Mitsuo Suzuki and Herman Karreman for advising and correcting the first draft of the manuscript; my wife Patricia for help with numerous details, and Mr. J. Wilson of Princeton University ...
For many time series data , we can say that the observations at any time t and t - 1 , that is , y , and y , -1 , are ... -1 , and may be with y , - ; for some i , with i = 2,3,4 , ... , we can exploit this correlation for forecasting .