Forecasting Economic Time SeriesAcademic Press, 10. 5. 2014 - Počet stran: 352 Economic Theory, Econometrics, and Mathematical Economics, Second Edition: Forecasting Economic Time Series presents the developments in time series analysis and forecasting theory and practice. This book discusses the application of time series procedures in mainstream economic theory and econometric model building. Organized into 10 chapters, this edition begins with an overview of the problem of dealing with time series possessing a deterministic seasonal component. This text then provides a description of time series in terms of models known as the time-domain approach. Other chapters consider an alternative approach, known as spectral or frequency-domain analysis, that often provides useful insights into the properties of a series. This book discusses as well a unified approach to the fitting of linear models to a given time series. The final chapter deals with the main advantage of having a Gaussian series wherein the optimal single series, least-squares forecast will be a linear forecast. This book is a valuable resource for economists. |
Obsah
1 | |
CHAPTER TWO SPECTRAL ANALYSIS | 45 |
CHAPTER THREE BUILDING LINEAR TIME SERIES MODELS | 76 |
CHAFFER FOUR THE THEORY OF FORECASTING | 120 |
CHAFER FIVE PRACTICAL METHODS FOR UNIVARIATE TIME SERIES FORECASTING | 151 |
CHAPTER SIX FORECASTING FROM REGRESSION MODELS | 187 |
CHAPTER SEVEN MULTIPLE SERIES MODELING AND FORECASTING | 216 |
CHAPTER EIGHT BUILDING MULTIPLE TIME SERIES FORECASTING MODELS | 235 |
CHAPTER NINE THE COMBINATION AND EVALUATION OF FORECASTS | 265 |
CHAPTER TEN FURTHER TOPICS | 297 |
317 | |
331 | |
335 | |
Běžně se vyskytující výrazy a sousloví
alternative analysis approach appropriate ARIMA ARMA model ARMA(p assumed assumption asymptotic autocovariance autoregressive model autoregressive process Box-Jenkins Chapter co-integrated coefficients combined forecast conditional considered correlation cost function covariance criterion denoted derived deterministic differencing discussed distribution econometric models employed equation estimated evaluation example exogenous variables exponential smoothing filter first-order autoregressive fitted model follows forecast errors formula frequency further given Granger Holt—Winters information set Kalman filter Lagrange multiplier lags least squares likelihood function linear matrix methods model building moving average moving average process Newbold nonlinear nonseasonal normally distributed null hypothesis observations obtained optimal forecast parameters partial autocorrelations particular polynomials possible practice predicted predictor problem procedure quantity random variables random walk regression residual autocorrelations sample autocorrelations seasonal component Section sequence series model specification spectral spectrum squared error stationary process stepwise autoregressive stochastic process suggest Table tion univariate variance vector white noise series zero zero-mean