GrunfeldGreene {systemfit} | R Documentation |
Panel data on 5 US firms for the years 1935-1954.
data("GrunfeldGreene")
A data frame containing 20 annual observations on 3 variables for 5 firms.
There exist several different versions of this data set,
and this version is considered incorrect
(see http://www.stanford.edu/~clint/bench/grunfeld.htm for details).
However, we provide this incorrect version to replicate the results
published in Theil (1971) and Greene (2003).
A correct version of this data set with 5 additional firms
is available in the Ecdat
package
(data set Grunfeld
).
Greene (2003), Appendix F, Data Sets Used in Applications, Table F13.1. http://pages.stern.nyu.edu/~wgreene/Text/econometricanalysis.htm (a subset of this data set is available in Theil (1971), p. 296).
Greene, W.H. (2003). Econometric Analysis, 5th edition. Prentice Hall, Upper Saddle River (NJ).
Grunfeld, Y. (1958). The Determinants of Corporate Investment, Unpublished Ph.D. Dissertation, University of Chicago.
Theil, Henri (1971). Principles of Econometrics, John Wiley & Sons, New York.
## Repeating the OLS and SUR estimations in Greene (2003, pp. 351) data( "GrunfeldGreene" ) library( plm ) GGPanel <- plm.data( GrunfeldGreene, c( "firm", "year" ) ) formulaGrunfeld <- invest ~ value + capital # OLS greeneOls <- systemfit( formulaGrunfeld, "OLS", data = GGPanel ) summary( greeneOls ) sapply( greeneOls$eq, function(x){return(summary(x)$ssr/20)} ) # sigma^2 # OLS Pooled greeneOlsPooled <- systemfit( formulaGrunfeld, "OLS", data = GGPanel, pooled = TRUE ) summary( greeneOlsPooled ) sum( sapply( greeneOlsPooled$eq, function(x){return(summary(x)$ssr)}) )/97 # sigma^2 # SUR greeneSur <- systemfit( formulaGrunfeld, "SUR", data = GGPanel, methodResidCov = "noDfCor" ) summary( greeneSur ) # SUR Pooled greeneSurPooled <- systemfit( formulaGrunfeld, "SUR", data = GGPanel, pooled = TRUE, methodResidCov = "noDfCor", residCovWeighted = TRUE ) summary( greeneSurPooled ) ## Repeating the OLS and SUR estimations in Theil (1971, pp. 295, 300) GrunfeldTheil <- subset( GrunfeldGreene, firm %in% c( "General Electric", "Westinghouse" ) ) GTPanel <- plm.data( GrunfeldTheil, c( "firm", "year" ) ) formulaGrunfeld <- invest ~ value + capital # OLS theilOls <- systemfit( formulaGrunfeld, "OLS", data = GTPanel ) summary( theilOls ) # SUR theilSur <- systemfit( formulaGrunfeld, "SUR", data = GTPanel, methodResidCov = "noDfCor" ) summary( theilSur )