出版社：Fundação Getulio Vargas, Escola de Pós-Graduação em Economia
摘要：In this paper, we propose a novel approach to econometric forecast-
ing of stationary and ergodic time series within a panel-data frame-
work. Our key element is to employ the bias-corrected average fore-
cast. Using panel-data sequential asymptotics we show that it is po-
tentially superior to other techniques in several contexts. In particular,it delivers a zero-limiting mean-squared error if the number of fore-
casts and the number of post-sample time periods is su¢ ciently large.
We also develop a zero-mean test for the average bias. Monte-Carlo
simulations are conducted to evaluate the performance of this new
technique in .nite samples. An empirical exercise, based upon data
from well known surveys is also presented. Overall, these results show
promise for the bias-corrected average forecast.
关键词：Panel-Data Econometrics, Pooling of Forecasts, Forecast
Combination Puzzle, Common Features
