Testing homogeneity of Japanese CPI forecasters

Journal of Forecasting (Impact Factor: 0.93). 01/2009; 29(5):435-441. DOI: 10.1002/for.1130
Source: RePEc


This paper investigates whether some forecasters consistently outperform others using Japanese CPI forecast data of 42 forecasters over the past 18 quarters. It finds that the accuracy rankings of 0, 1, 2, and 5-month forecasts are significantly different from those that might be expected when all forecasters had equal forecasting ability. Moreover, their rankings of the relative forecast levels are also significantly different from a random one. Copyright © 2009 John Wiley & Sons, Ltd.

5 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper examines interest rate forecasts made for the period 1982–90 and examines three issues: (1) Is there a general agreement among analysts about the level of interest rates six months in the future? (2) Are all the forecasters equally good? (3) Are the forecasts valuable to prospective users? We use distributions of the cross-sections of forecasts, Friedman's statistic for analysis of variance by rank, and tests of independence between forecasts and outcomes to examine these questions. We conclude that there usually was a consensus among analysts, that there was no significant difference in the ability to forecast short-term rates but there was a difference with respect to the long-term predictions, and that these forecasts were not significantly better than random walk forecasts.
    Journal of Forecasting 09/1996; 15(5):385 - 394. DOI:10.1002/(SICI)1099-131X(199609)15:5<385::AID-FOR627>3.0.CO;2-B · 0.93 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    Journal of Accounting Research 01/1990; 28(2). DOI:10.2307/2491151 · 2.38 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Distribution-free tests and multiple comparison procedures are considered for randomized block designs. A general Friedman-type statistic is defined that is usable in almost any block design with arbitrary missing data structure. For balanced incomplete block designs a multiple comparison procedure and exact tables are given for a statistic equivalent to one proposed by Durbin (1951). Finally, for designs with missing observations on only one treatment, we present exact tables for the test statistic, and we consider a multiple comparison procedure.
    Technometrics 05/1981; 23(2):171-177. DOI:10.1080/00401706.1981.10486261 · 1.81 Impact Factor


5 Reads
Available from