MARC Record from Library of Congress

LEADER: 02164cam a22003137a 4500
001 2005301035
003 DLC
005 20051230133643.0
007 cr |||||||||||
008 051230s2005 mou sb f000 0 eng
010 $a 2005301035
040 $aDLC$cDLC
050 00 $aHB1
100 1 $aClark, Todd E.
245 10 $aApproximately normal tests for equal predictive accuracy in nested models$h[electronic resource] /$cTodd E. Clark and Kenneth D. West.
260 $aKansas City [Mo.] :$bResearch Division, Federal Reserve Bank of Kansas City,$c[2005]
490 1 $aRWP ;$v05-05
538 $aSystem requirements: Adobe Acrobat Reader.
538 $aMode of access: World Wide Web.
500 $aTitle from PDF file (viewed on Dec. 30, 2005).
500 $a"November 2005."
530 $aAlso available in print.
504 $aIncludes bibliographical references.
520 3 $a"Forecast evaluation often compares a parsimonious null model to a larger model that nests the null model. Under the null that the parsimonious model generates the data, the larger model introduces noise into its forecasts by estimating parameters whose population values are zero. We observe that the mean squared prediction error (MSPE) from the parsimonious model is therefore expected to be smaller than that of the larger model. We describe how to adjust MSPEs to account for this noise. We propose applying standard methods (West (1996)) to test whether the adjusted mean squared error difference is zero. We refer to nonstandard limiting distributions derived in Clark and McCracken (2001, 2005a) to argue that use of standard normal critical values will yield actual sizes close to, but a little less than, nominal size. Simulation evidence supports our recommended procedure."
653 $aForecast evaluation ;$aCausality ;$aNested models
700 1 $aWest, Kenneth D.$q(Kenneth David)
710 2 $aFederal Reserve Bank of Kansas City.$bResearch Division.
830 0 $aResearch working paper (Federal Reserve Bank of Kansas City : Online) ;$v05-05.
856 42 $3Abstract page with link to full-text PDF report$uhttp://www.kc.frb.org/PUBLICAT/RESWKPAP/RWP05-05.htm