search for: epsihf

Displaying 3 results from an estimated 3 matches for "epsihf".

2005 Feb 23
1
H-F corr.: covariance matrix for interaction effect
...S) # n - number of measurements (i.e. number of rows in original matrix) epsi.GG.HF <- function (S,k,n) { D <- (k^2 * (mean(S) - mean(diag(S)))^2) N1 <- sum(S^2) N2 <- 2 * k * sum(apply(S, 1, mean)^2) N3 <- k^2 * mean(S)^2 epsiGG <- D / ((k - 1) * (N1 - N2 + N3)) epsiHF <- (n * (k-1) * epsiGG - 2) / ((k-1) * ((n-1) - (k-1)*epsiGG)) c(epsiGG,epsiHF) } # three factors, facROI,facCond,facSubj # facROI,facCond are main effects, facSubj is "repeatedness" # G-G and H-F corrections for a main effect # we do the gghf stuff for the ROI, which means ROIs i...
2005 Feb 23
1
H-F corr.: covariance matrix for interaction effect
...S) # n - number of measurements (i.e. number of rows in original matrix) epsi.GG.HF <- function (S,k,n) { D <- (k^2 * (mean(S) - mean(diag(S)))^2) N1 <- sum(S^2) N2 <- 2 * k * sum(apply(S, 1, mean)^2) N3 <- k^2 * mean(S)^2 epsiGG <- D / ((k - 1) * (N1 - N2 + N3)) epsiHF <- (n * (k-1) * epsiGG - 2) / ((k-1) * ((n-1) - (k-1)*epsiGG)) c(epsiGG,epsiHF) } # three factors, facROI,facCond,facSubj # facROI,facCond are main effects, facSubj is "repeatedness" # G-G and H-F corrections for a main effect # we do the gghf stuff for the ROI, which means ROIs i...
2007 Jun 24
2
ANOVA non-sphericity test and corrections (eg, Greenhouse-Geisser)
...Feldt correction # The Greenhouse-Geisser epsilon tends to underestimate # epsilon when epsilon is greater than 0.70 (Stevens, 1990). # An estimated e=0.96 may be actually 1. Huynh-Feldt # correction is less conservative. The Huynh-Feldt # epsilon is calculated from the Greenhouse-Geisser epsilon, epsiHF <- (n * (k-1) * epsi - 2) / ((k-1) * ((n-1) - (k-1)*epsi)) PepsiHF <- 2 * (1 - pf(Fvalue, df1=epsiHF*(k-1), df2=epsiHF*(k-1)*(n-1) ) ) # MANOVA (and multivariate tests) may be better if the # Greenhouse-Geisser and the Huynh-Feldt corrections do not agree, # which may happen when epsilon dr...