Use this to set options and select the output to be generated by Hierarchical Generalized Linear Model analysis.
This specifies which items of output are to be produced by the analysis.
|Model||Details of the model that has been fitted|
|Deviance||Estimates of the deviances|
|Monitoring||Monitoring information at each iteration|
|Fitted values||Table with unit number, response variable, fitted values, residuals and leverages|
|Estimates (fixed model)||Estimates of the parameters in the fixed model|
|Estimates (random model)||Estimates of the parameters in the random model|
|Estimates (dispersion)||Estimates of the parameters in the dispersion models (if any)|
|Likelihood statistics||Likelihood statistics for assessing the models|
|Wald tests (fixed model)||Wald statistics for the terms in the fixed model|
Estimates of censored observations
|When censoring is selected for a Poisson-log link model, this prints the estimates of censored observations|
Controls whether the dispersion parameter for the variance of the response is estimated from the residual mean square of the fitted model, or fixed at a given value. The dispersion parameter (fixed or estimated) is used when calculating standard errors and standardized residuals. In models with the binomial or Poisson distributions, the dispersion should be fixed at 1 unless a heterogeneity parameter is to be estimated.
Estimate constant term
Specifies whether to include a constant in the HGLM fixed model. If the constant is not included, the intercept term will be zero.
Maximum number of iterations
|GLM||maximum number of iterations of the hierarchical generalized linear algorithm|
|Mean dispersion model||maximum number of iterations in the fitting of the mean and dispersion models|
|Censored E-M||maximum number of iterations in the fitting censored observations with the Tobit expectation-maximization algorithm|
|HGLM||maximum change in hierarchical generalized linear model parameters between iterations|
|Aitken extrapolation||maximum size of ratio of the original to the new estimates allowed in Aitken extrapolation|
|Censored E-M||maximum change in estimated censored observations with the Tobit expectation-maximization algorithm|
Order of Laplace approximation in mean model
This can be used to set the order of the Laplace approximation to use in the estimation of the mean model.
Order of Laplace approximation in dispersion model
This can be used to set the order of the Laplace approximation involved in the estimation of the dispersion components of the model. This is appropriate for generalized linear mixed models with the binomial or Poisson distributions, where use of Laplace order 0 can lead to serious downwards bias. Caution must be used when setting an order of 2 as the analysis can take a long time to run.
A fixed model in an HGLM can be modified to take account of a fixed contribution to the linear effects for each unit, supplied in a variate referred to as the offset.
This lets you specify a variate of prior for the weights for the fixed model.
Factorial limit on fixed model terms
For the fixed model you can control the factorial limit on model terms to be generated when you use model-formula operators like *. The default is to include all interactions, up to those involving nine variates or factors.
|Run||Save options and close dialog.|
|Cancel||Close dialog without saving the options.|
|Defaults||Reset the options to their default settings.|
|Clear||Clear all fields and list boxes.|
|Help||Open the Help topic for this dialog.|
- HGLM menu for fitting HGLMs.
- HGLM Save Options.
- HGLM Further Output.
- HGLM Predictions.
- HGLM Model Checking for residual plots.
- Fitted model menu for plotting the fit of an HGLM.
- HGLM Likelihood Tests for Fixed Terms dialog.
- HGLM Likelihood Tests for Random Terms dialog.
- HGANALYSE and
- HGTOBITPOISSON procedures.