Mizrach robust forecast comparison statistic
mizrachl2 pred1 pred2 [if exp] [in range] [,k(lags) ]
mizrachl2 is for use with time-series data. You must tsset your data before using mizrachl2; see help tsset.
varname may contain time-series operators; see help varlist.
mizrachl2 calculates a forecast comparison test proposed by Mizrach (1995). Given two competing predictions, a positive definite, robust statistic for comparing the predictions' accuracy may be calculated.
k specifies the maximum order of the lag to be used in calculating the long-run variance of the difference series from its autocovariance function. If it is not provided, the maximum lag order will be calculated as a function of the sample size.
. use http://fmwww.bc.edu/ec-p/data/macro/barclaymonthly.dta,clear
. mizrachl2 bbjpy1f bbjpy2f
. mizrachl2 bbjpy1f bbjpy2f, k(10)
. mizrachl2 D.bbjpy1f D.bbjpy2f if tin(1987m12,1996m12), k(4)
Bruce Mizrach, "Forecast Comparison in L2", Rutgers University Working Paper 1995-24
This routine is a translation of a Fortran program kindly provided by Bruce Mizrach. Remaining errors are my own.
Christopher F Baum, Boston College, USA, email@example.com
On-line: help for predict; help for dmariano (if installed)