¥3dRegAna: Other
statistics
HHow well does model fit
data?
åR2 (coefficient of multiple determination)
is the proportion of the variance in the data accounted for by the model 0 ² R2 ² 1.
åi.e. if R2 = 0.26 then 26% of the dataÕs variation
about their mean is accounted for by the model. So this might indicate the model,
while significant might not be that useful.
íHaving said that, you should
consider R2 relative to the maximum it can achieve given the pure error which
cannot be modeled. [read Draper & Smith, chapter 2].
HAre individual parameters bk significant?
åt-statistic is
calculated for each parameter
åhelps identify parameters that can be discarded to simplify the
model
HR2 and t-statistic are computed for
full (not reduced) model
¥