tailieunhanh - Handbook of Economic Forecasting part 78

Handbook of Economic Forecasting part 78. Research on forecasting methods has made important progress over recent years and these developments are brought together in the Handbook of Economic Forecasting. The handbook covers developments in how forecasts are constructed based on multivariate time-series models, dynamic factor models, nonlinear models and combination methods. The handbook also includes chapters on forecast evaluation, including evaluation of point forecasts and probability forecasts and contains chapters on survey forecasts and volatility forecasts. Areas of applications of forecasts covered in the handbook include economics, finance and marketing | 744 . Pesaran and M. Weale The model is then estimated using a Kalman filter technique finding that the time-series model is preferred to the standard model. Cunningham Smith and Weale 1998 and Mitchell Smith and Weale 2002 relate survey responses to official data by regressing the proportions of firms reporting rises and falls on the official data. Cunningham Smith and Weale 1998 however take the view that the survey data represent some transformation of the underlying latent variable with an additional error term added on arising for perception and measurement reasons. For this reason it may seem more appropriate to estimate regression equations which explain observed proportions Ue 1 and Det 1 or Ut and Dt rather than explaining output by the survey aggregates. This means that estimates of the variable represented by the survey have to be derived by inverting each regression equation. Since the number of independent regression equations is equal to the number of categories less one there are this number of separate estimates of the variable of interest produced. Since however the covariance of the vector of these distinct estimates can be estimated from the standard properties of regression equations it is possible to produce a variance-weighted mean of the different estimates to give a best estimate of the variable of interest Stone Champernowne and Meade 1942 . Mitchell Smith and Weale 2002 extend this technique using the CBI survey data. Instead of explaining the two survey proportions the proportion reporting or expecting a rise in output and the proportion reporting or expecting a fall they look at the proportions reporting expecting rises or falls as proportions of those who had re-ported expected rises no change or falls in the previous This creates a system of six equations which can be estimated in the same way. Mitchell Smith and Weale describe this as a semi-disaggregate approach. They find evidence suggesting that the thresholds which .