tailieunhanh - Handbook of Economic Forecasting part 90

Handbook of Economic Forecasting part 90. 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 | 864 . Andersen et al. Moreover combining a number of volatility forecasts may be preferable to choosing a single best forecast. The general topic of forecast combination is discussed in detail in Chapter 4 by Timmermann in this Handbook. Volatility forecast combination has been found to work well in practice by Hu and Tsoukalas 1999 . Further discussion of volatility forecasting and forecast evaluation based on realized volatility measures can be found in Andersen and Bollerslev 1998a Andersen Boller-slev and Meddahi 2004 2005 and Patton 2005 . Andersen et al. 1999 Ait-Sahalia Mykland and Zhang 2005 Bandi and Russel 2003 2004 Bollen and Inder 2002 Curci and Corsi 2004 Hansen and Lunde 2004b Martens 2003 and Zhang Mykland and Ait-Sahalia 2005 all analyze the important choice of sampling frequency and or the use of various sub-sampling and other corrective procedures in the practical construction of unbiased and efficient realized volatility measures. Alizadeh Brandt and Diebold 2002 discuss the relative merits of realized and range-based volatility. For early work on the properties of range-based estimates see Feller 1951 and Parkinson 1980 . Testing for normality of the transformed Probability Integral Transform PIT variable can be done in numerous ways. A couple of interesting recent procedures for testing dynamic models for correct distributional assumptions taking into account the parameter estimation error uncertainty are given by Bontemps and Meddahi 2005 and Duan 2003 . Several important topics were not explicitly discussed in this section. In the general forecasting area they include covariance and correlation forecast evaluation see . Brandt and Diebold 2006 as well as related multivariate density forecast evaluation see . Diebold Hahn and Tay 1999 . In the area of financial forecast applications we did not discuss the evaluation of time-varying betas see . Ghysels 1998 volatility-based asset allocation see . Fleming Kirby and Ostdiek 2001 .