๐”– Bobbio Scriptorium
โœฆ   LIBER   โœฆ

On the limitations of comparing mean square forecast errors: Comment

โœ Scribed by Stephen K. McNees

Publisher
John Wiley and Sons
Year
1993
Tongue
English
Weight
205 KB
Volume
12
Category
Article
ISSN
0277-6693

No coin nor oath required. For personal study only.

โœฆ Synopsis

An algebraic relationship between mean square error comparisons and encompassing tests is provided by a contrast of two of the earliest regression equations for comparing forecasting formulas. Hoel (1947) proposed a regression equation of precisely the form given by CH (equation (44)), namely,

where y, y1, and y2 are the actual and predicted values from two forecasting equations. Williams and Moot (1953) proposed a regression of the form ( 5 ) which Williams (1959) showed was related to Wilks's (1946) homogeneity of variance test and hence a test of the hypothesis y = 0 (equal weights attached to each forecast) is a type of mean square error comparison which takes sampling variability into account. The relationship between and y is given by y = P -1/2 so that if P = 0 so that y1 encompasses yz, y < 0 which indicates that y~ has a smaller MSFE than yz.

Yt ( Y l +

๐Ÿ“œ SIMILAR VOLUMES

On the limitations of comparing mean squ
On the limitations of comparing mean square forecast errors: Comment
โœ Clive W. J. Granger ๐Ÿ“‚ Article ๐Ÿ“… 1993 ๐Ÿ› John Wiley and Sons ๐ŸŒ English โš– 149 KB

Meese, R. A. and Rogoff, K., 'Empirical exchange rate models of the seventies: do they fit out of sample?' Journal of International Economics, 14 (1983a). 1, 3-24. Meese, R. A. and Rogoff, K. 'The out-of-sample failure of empirical exchange rate models: sampling error or misspecification?', Chapter

On the limitations of comparing mean squ
On the limitations of comparing mean square forecast errors: Comment
โœ Richard A. Meese ๐Ÿ“‚ Article ๐Ÿ“… 1993 ๐Ÿ› John Wiley and Sons ๐ŸŒ English โš– 189 KB

See Chinn and Meese (1992) for evidence and justification of constrained parameter values. They use x(r) = [ s ( t )log(rn(r)/rn\*(r)) + 0.75 log(y(f)/y\*(t)) -4.5(i(f) -i\*(r))l, where rn, y , and i are domestic money supply, real income, and interest rate proxies, respectively. Foreign values of t

On the limitations of comparing mean squ
On the limitations of comparing mean square forecast errors: Comment
โœ Peter Schmidt ๐Ÿ“‚ Article ๐Ÿ“… 1993 ๐Ÿ› John Wiley and Sons ๐ŸŒ English โš– 234 KB

one-step forecasts is, of course, simply the analogy of many in-sample tests of model specification. In this context, it is amusing that much recent applied econometric work has, in effect, exploited fitted models to make inference about economic systems, as the forecast horizon approaches injinity!