Next: betalimits() Up: Regression Macros Help File Previous: Regression Macros Help File   Contents


anovapred(a,b,...), a, b, ... all the factors in STRMODEL

Keywords: glm, anova, standard error, confidence limits, prediction limits
anovapred(a,b,...), where a, b, ... are all the factors in the most
recent GLM model, computes the fitted (predicted) value, the standard
error of estimation, and the standard error of prediction for each
cell. The result is a structure with components 'estimate', 'SEest' and
'SEpred', each of which is a vector, matrix, or array with dimensions
derived from the number of levels of a, b, ....  It uses side effect

When the most recent GLM model was manova() with a p-dimensional
dependent variable, each component will have an extra dimension of size

When the most recent GLM model included variates (non-factors), or when
you do not include all factors in the argument list, the results will
probably be wrong, although no warning message will be printed.

anovapred is implemented as a pre-defined macro.

See also predtable(), regpred(), 'glm'.

Gary Oehlert 2003-01-15