glmmnode properties | IBM Cloud Pak for Data as a Service

glmmnode properties

Last updated: Jan 17, 2024

glmmnode properties

GLMM node icon A generalized linear mixed model (GLMM) extends the linear model so that the target can have a non-normal distribution, is linearly related to the factors and covariates via a specified link function, and so that the observations can be correlated. GLMM models cover a wide variety of models, from simple linear regression to complex multilevel models for non-normal longitudinal data.

Table 1. glmmnode properties
`glmmnode` Properties	Values	Property description
`residual_subject_spec`	structured	The combination of values of the specified categorical fields that uniquely define subjects within the data set
`repeated_measures`	structured	Fields used to identify repeated observations.
`residual_group_spec`	[field1 ... fieldN]	Fields that define independent sets of repeated effects covariance parameters.
`residual_covariance_type`	`Diagonal` `AR1` `ARMA11` `COMPOUND_SYMMETRY` `IDENTITY` `TOEPLITZ` `UNSTRUCTURED` `VARIANCE_COMPONENTS`	Specifies covariance structure for residuals.
`custom_target`	flag	Indicates whether to use target defined in upstream node (`false`) or custom target specified by `target_field` (`true`).
`target_field`	field	Field to use as target if `custom_target` is `true`.
`use_trials`	flag	Indicates whether additional field or value specifying number of trials is to be used when target response is a number of events occurring in a set of trials. Default is `false`.
`use_field_or_value`	`Field` `Value`	Indicates whether field (default) or value is used to specify number of trials.
`trials_field`	field	Field to use to specify number of trials.
`trials_value`	integer	Value to use to specify number of trials. If specified, minimum value is 1.
`use_custom_target_reference`	flag	Indicates whether custom reference category is to be used for a categorical target. Default is `false`.
`target_reference_value`	string	Reference category to use if `use_custom_target_reference` is `true`.
`dist_link_combination`	`Nominal` `Logit` `GammaLog` `BinomialLogit` `PoissonLog` `BinomialProbit` `NegbinLog` `BinomialLogC` `Custom`	Common models for distribution of values for target. Choose `Custom` to specify a distribution from the list provided by`target_distribution`.
`target_distribution`	`Normal` `Binomial` `Multinomial` `Gamma` `Inverse` `NegativeBinomial` `Poisson`	Distribution of values for target when `dist_link_combination` is `Custom`.
`link_function_type`	`Identity` `LogC` `Log` `CLOGLOGLogit` `NLOGLOGPROBIT` `POWER` `CAUCHIT`	Link function to relate target values to predictors. If `target_distribution` is `Binomial` you can use any of the listed link functions. If `target_distribution` is `Multinomial` you can use `CLOGLOG`, `CAUCHIT`, `LOGIT`, `NLOGLOG`, or `PROBIT`. If `target_distribution` is anything other than `Binomial` or `Multinomial` you can use `IDENTITY`, `LOG`, or `POWER`.
`link_function_param`	number	Link function parameter value to use. Only applicable if `normal_link_function` or `link_function_type` is `POWER`.
`use_predefined_inputs`	flag	Indicates whether fixed effect fields are to be those defined upstream as input fields (`true`) or those from `fixed_effects_list` (`false`). Default is `false`.
`fixed_effects_list`	structured	If `use_predefined_inputs` is `false`, specifies the input fields to use as fixed effect fields.
`use_intercept`	flag	If `true` (default), includes the intercept in the model.
`random_effects_list`	structured	List of fields to specify as random effects.
`regression_weight_field`	field	Field to use as analysis weight field.
`use_offset`	`Noneoffset_valueoffset_field`	Indicates how offset is specified. Value `None` means no offset is used.
`offset_value`	number	Value to use for offset if `use_offset` is set to `offset_value`.
`offset_field`	field	Field to use for offset value if `use_offset` is set to `offset_field`.
`target_category_order`	`AscendingDescendingData`	Sorting order for categorical targets. Value `Data` specifies using the sort order found in the data. Default is `Ascending`.
`inputs_category_order`	`AscendingDescendingData`	Sorting order for categorical predictors. Value `Data` specifies using the sort order found in the data. Default is `Ascending`.
`max_iterations`	integer	Maximum number of iterations the algorithm will perform. A non-negative integer; default is 100.
`confidence_level`	integer	Confidence level used to compute interval estimates of the model coefficients. A non-negative integer; maximum is 100, default is 95.
`degrees_of_freedom_method`	`FixedVaried`	Specifies how degrees of freedom are computed for significance test.
`test_fixed_effects_coeffecients`	`ModelRobust`	Method for computing the parameter estimates covariance matrix.
`use_p_converge`	flag	Option for parameter convergence.
`p_converge`	number	Blank, or any positive value.
`p_converge_type`	`AbsoluteRelative`
`use_l_converge`	flag	Option for log-likelihood convergence.
`l_converge`	number	Blank, or any positive value.
`l_converge_type`	`AbsoluteRelative`
`use_h_converge`	flag	Option for Hessian convergence.
`h_converge`	number	Blank, or any positive value.
`h_converge_type`	`AbsoluteRelative`
`max_fisher_step`	integer
`sing_tolerance`	number
`use_model_name`	flag	Indicates whether to specify a custom name for the model (`true`) or to use the system-generated name (`false`). Default is `false`.
`model_name`	string	If `use_model_name` is `true`, specifies the model name to use.
`confidence`	`onProbabilityonIncrease`	Basis for computing scoring confidence value: highest predicted probability, or difference between highest and second highest predicted probabilities.
`score_category_probabilities`	flag	If `true`, produces predicted probabilities for categorical targets. Default is `false`.
`max_categories`	integer	If `score_category_probabilities` is `true`, specifies maximum number of categories to save.
`score_propensity`	flag	If `true`, produces propensity scores for flag target fields that indicate likelihood of "true" outcome for field.
`emeans`	structure	For each categorical field from the fixed effects list, specifies whether to produce estimated marginal means.
`covariance_list`	structure	For each continuous field from the fixed effects list, specifies whether to use the mean or a custom value when computing estimated marginal means.
`mean_scale`	`OriginalTransformed`	Specifies whether to compute estimated marginal means based on the original scale of the target (default) or on the link function transformation.
`comparison_adjustment_method`	`LSDSEQBONFERRONISEQSIDAK`	Adjustment method to use when performing hypothesis tests with multiple contrasts.
`use_trials_field_or_value`	`"field"` `"value"`
`residual_subject_ui_spec`	array	Residual subject specification: The combination of values of the specified categorical fields should uniquely define subjects within the dataset. For example, a single Patient ID field should be sufficient to define subjects in a single hospital, but the combination of Hospital ID and Patient ID may be necessary if patient identification numbers are not unique across hospitals.
`repeated_ui_measures`	array	The fields specified here are used to identify repeated observations. For example, a single variable Week might identify the 10 weeks of observations in a medical study, or Month and Day might be used together to identify daily observations over the course of a year.
`spatial_field`	array	The variables in this list specify the coordinates of the repeated observations when one of the spatial covariance types is selected for the repeated covariance type.