GLZ Results
Click the OK button in the
GLZ
Quick Specs dialog box to display the GLZ
Results dialog box, which contains five tabs:
Summary,
Resid.
1, Resid.
2, Means, and Report. This dialog box
will also be displayed when you click the OK
(Run) button in either the GLZ
Analysis Wizard Between Design dialog box,
the GLZ
Analysis Wizard Extended Options
dialog box, or the GLZ
Analysis Syntax Editor.
The Summary tab
contains options to produce
summaries of the main results, for example, tests of all effects, parameter
estimates, overall goodness of fit tests, descriptive statistics, etc.
The Resid.
1 tab contains options to produce predicted values as well
as raw residuals, studentized residuals, Pearson
residuals, etc. The Resid. 2 tab contains options to
produce various advanced plots of predicted and residual statistics. The
Means tab contains options
to produce 1) observed unweighted means, 2) observed weighted means, and
3) predicted means (given the current model). Finally, the Report tab is used to send
results to a report.
Modify.
Click the Modify button to display
the previous dialog box for the respective analysis (see Methods
for Specifying Designs). You will then be able to modify the current
analysis.
Close.
Click the Close button to close
the current results dialog and return to the Generalized Linear/Nonlinear Models Startup Panel.
Options.
Click the Options button to display
the Options menu.
By Group.
Click the By Group button to
display the By
Group specification dialog box.
Note: Results for stepwise or best-subset regression.
Unlike in the stepwise or best-subset results in General Regression
Models (GRM), the results that can be reviewed from this results
dialog always pertain to the full model, regardless of which effects were
selected for inclusion during the model building procedure. The reason
for this is that, unlike in GRM, the relationship between predictors,
and their interactive effects (e.g., two predictors masking the effects
of a third) are often much more complex. Also, unlike in GRM, because
of the manner in which the p1, enter
and p2, remove probabilities
are determined (in forward stepwise selection, the score
statistic is used to select new (significant) effects; while the Wald statistic
is used during backward steps), the Stepwise
(forward, backward) methods may result in the repetitive selection
and removal of one or more predictors. Therefore, the stepwise results
can be reviewed separately in this dialog, via the Model
building button on the Summary
tab. If, after comparing the overall model (with all effects)
with the one suggested by the model building procedure, you decide to
further evaluate the latter model, use the Make
model button on the Summary
tab to transfer that model to the Quick Specs
dialog box or the GLZ
Analysis Syntax Editor, and then fit that model to the data.
Note: Models that are not full-rank (e.g., overparameterized
models). When redundant columns are detected during the evaluation
of the design
matrix, some difficulties arise when computing the Wald
statistic for the overall model (see Summary
of all effects option on the Summary tab), and when
attempting to compute Type 3 LR (likelihood-ratio)
tests of effects on the Summary
tab (see also the GLM
topic Six
types of sums of squares). Specifically, because of the redundancy
of some of the parameters in the model, independent tests of effects,
controlling for all other parameters in the model (not belonging to the
effect under consideration) cannot be computed. Therefore, the Summary
of all effects and Type 3 LR
test buttons will not be available (on the Summary tab) in that
case.
Note: Reference level for categorical dependent
(response) variables. The last category (level) that is specified
for a categorical dependent
(response) variable will be the reference category for the comparisons
with the other categories. So, for example, if a multinomial dependent
(response) variable with k = 3 levels is analyzed, the k-1 = 2 parameters
for each predictor (effect column) pertain to the comparison of 1) the
first level with the last level, and 2) the second level with the last
level of the dependent (response) variable.
Note: Overdispersion parameter for models with discrete/categorical
responses. Poisson, Binomial,
Multinomial,
and Ordinal
multinomial distributions all have a default dispersion parameter
of 1. The data, however, may exhibit greater variability than this allows.
You can select the Overdispersion
check box on the Summary
tab and select either the Pearson
Chi-square or Deviance
option button. This enables you to specify a value for the dispersion
parameter, φ, with the scale parameter σ= √φ.
When you specify Pearson Chi-square,
STATISTICA uses the Pearson chi-square
statistic divided by its degrees of freedom as an estimate of φ.
When you specify Deviance,
STATISTICA uses the deviance
divided by its degrees of freedom as an estimate of φ.
Changing the overdispersion parameter will
affect the computation of the estimated parameter covariance matrix, the
model likelihood, and all related statistics (e.g., standard errors, prediction
errors, etc). For details, refer to McCullagh and Nelder, 1989.
See also, GLZ
Index.