Frequently Asked Questions

550 questions found.

Page 10 / 11

The copulaFit action estimates the parameters for a specified copula type, providing a way to model the dependence structure between multiple variables....
Read full answer See technical action

You can estimate several types of copulas by specifying the `copulatype` parameter. The available types are 'CLAYTON', 'FRANK', 'GUMBEL', 'NORMAL', and 'T'....
Read full answer See technical action

The `method` parameter allows you to choose the estimation technique. 'MLE' stands for Maximum Likelihood Estimation, which is the default. 'CAL' stands for the calibration method, which is typically faster but might be less accurate....
Read full answer See technical action

The input data is specified using the required `table` parameter, which points to the CAS table containing the variables to be analyzed....
Read full answer See technical action

The `var` parameter is a required argument where you provide a list of the variable names from the input table that you want to use for fitting the copula....
Read full answer See technical action

You can use the `store` parameter to save the model's properties and fit results into a CAS item store. You just need to provide a name for the store....
Read full answer See technical action

The `marginals` parameter specifies the marginal distribution of the individual variables. You can set it to 'EMPIRICAL' (the default) to use the empirical distribution from the data, or 'UNIFORM' if the data are already transformed to uniform marginals....
Read full answer See technical action

For fitting a t-copula, you can provide a Pearson correlation matrix using the `corrtable` parameter or a Kendall's tau correlation matrix using the `KendallCorrtable` parameter. You can also specify an initial value for the degrees of freedom using the `df` parameter....
Read full answer See technical action

Yes, the `plot` parameter allows you to generate various diagnostic plots, such as scatter plots of the data (`scatter`), Kendall's tau plots (`kendall`), and tail dependence plots (`tail`)....
Read full answer See technical action

You can use the `outpseudo` parameter to specify an output CAS table where the pseudo-samples, transformed to have uniform marginal distributions, will be saved....
Read full answer See technical action

The copulaSimulate action simulates data from a specified copula model....
Read full answer See technical action

The 'define' parameter specifies the relevant information about the copula that is used for the simulation....
Read full answer See technical action

You can specify the following copula types: "CLAYTON", "FRANK", "GUMBEL", "NORMAL", or "T". The default is "NORMAL"....
Read full answer See technical action

For elliptical copulas, you can specify the Pearson correlation matrix using the 'corrtable' parameter, the Kendall correlation matrix with 'KendallCorrtable', or the Spearman correlation matrix with 'SpearmanCorrtable' for Gaussian copulas....
Read full answer See technical action

The 'df' parameter specifies the degrees of freedom and is only applicable for T copulas. Its minimum value is 1....
Read full answer See technical action

Use the 'theta' parameter (aliased as 'alpha') to specify the parameter value for Archimedean copulas....
Read full answer See technical action

The 'ndraws' parameter specifies the number of observations to generate for the simulation. The default value is 10,000....
Read full answer See technical action

You can save the simulation results in two ways: use the 'outuniform' parameter to get the result in uniform margins, or use the 'outempirical' parameter to get the result using empirical margins....
Read full answer See technical action

The 'seed' parameter specifies the seed for generating random numbers for the simulation, allowing for reproducibility. The default value is 0....
Read full answer See technical action

Yes, you can use the 'restore' parameter to specify an item store that contains the properties and estimates of a previously fitted model. The action will use this restored information for the simulation....
Read full answer See technical action

The 'margApproxOpts' parameter controls the approximation. You can choose the 'algorithm' ('BIN' or 'SORT'), the 'interpolation' method ('LINEAR', 'MONOCUBIC', or 'STEP'), the maximum number of iterations ('maxiters'), and the error tolerance ('sampletol')....
Read full answer See technical action

The copulaViewStore action is used to display the model and estimates that have been preserved in a specified item store....
Read full answer See technical action

The 'instore' parameter is required. It is used to specify the item store from which to restore the model properties and final estimates....
Read full answer See technical action

You can use the 'viewOptions' parameter to control the output. It accepts several boolean flags such as 'all', 'minimal', 'finalEstimates', 'correlations', 'fitModelSummary', and 'modelDefinition' to customize which reports are displayed....
Read full answer See technical action

Yes, the 'outputTables' parameter can be used to specify a list of display tables that you want to save as CAS output tables....
Read full answer See technical action

The 'display' parameter allows you to specify a list of display tables that you want the action to generate. If this parameter is omitted, the action creates all available tables by default....
Read full answer See technical action

The copyModelExternal action copies a model from a CAS table to an external database....
Read full answer See technical action

The required parameters for the copyModelExternal action are `externalOptions`, `modelName`, and `modelTable`....
Read full answer See technical action

You must specify the source model using the `modelTable` parameter, which is a CAS table containing the model, and the `modelName` parameter, which identifies the specific model within that table....
Read full answer See technical action

The destination is defined using the `externalOptions` parameter. You must specify the `extType` (e.g., 'DATABRICKS', 'FILESYSTEM', 'HADOOP', 'SINGLESTORE', 'SYNAPSE', 'TERADATA') and provide the relevant parameters for that specific external type....
Read full answer See technical action

Yes, you can overwrite an existing model by setting the `replace` parameter within `modelOptions` to TRUE. This is the default behavior....
Read full answer See technical action

The supported values for the `extType` parameter are 'DATABRICKS', 'FILESYSTEM', 'HADOOP', 'SINGLESTORE', 'SYNAPSE', and 'TERADATA'....
Read full answer See technical action

When using 'TERADATA', you must provide the Teradata model table name via the `modelTable` parameter within `externalOptions`. You may also need to provide connection details such as `server`, `database`, `username`, and `password` or an `authenticationDomain`....
Read full answer See technical action

For `extType='FILESYSTEM'`, the required parameter is `modelDir`, which specifies the root folder where the model directory will be created....
Read full answer See technical action

The network.core action is used to calculate the core decomposition of a graph. This process identifies dense subgraphs, which can be useful for understanding the structure of the network....
Read full answer See technical action

A k-core of a graph is a maximal subgraph in which every node has a degree of at least k. The core decomposition of a graph is a nested set of these k-cores....
Read full answer See technical action

The 'direction' parameter specifies whether the graph should be treated as 'DIRECTED' or 'UNDIRECTED'. In an undirected graph, links are bidirectional, while in a directed graph, links have a specific orientation from a source to a sink node. The default is 'UNDIRECTED'....
Read full answer See technical action

You can set the 'distributed' parameter to 'TRUE'. This option enables the use of a distributed graph processing algorithm, allowing the analysis to be performed across multiple machines....
Read full answer See technical action

The core action can generate two main output tables: 'outNodes', which contains the core number for each node in the graph, and 'outLinks', which contains the links that form the resulting core subgraphs....
Read full answer See technical action

The 'ProblemSummary' result table provides a basic summary of the input graph. The 'SolutionSummary' result table contains a basic summary of the algorithm's execution, including the solution status....
Read full answer See technical action

The simple.correlation action is used to compute Pearson product-moment correlations....
Read full answer See technical action

To compute Cronbach's coefficient alpha, you need to set the 'alpha' parameter to TRUE....
Read full answer See technical action

You can use the 'listwiseDelMiss' parameter. When set to TRUE, it applies listwise deletion to observations with missing values. By default, this is FALSE....
Read full answer See technical action

The 'varianceDivisor' parameter (alias 'varDef') is used for this purpose. It accepts the following values: 'DF' (degrees of freedom, default), 'N' (number of observations), 'WDF' (sum of weights minus one), or 'WEIGHT' (sum of weights)....
Read full answer See technical action

Yes. You can obtain a variance/covariance matrix by setting 'covariance' to TRUE, a corrected sum of squares and cross-products with 'csscp', and a sum of squares and cross-products with 'sscp'. Univariate descriptive statistics are also generated by default via the 'descriptiveStats' parameter....
Read full answer See technical action

You can use the 'casOut' or 'pearsonOut' parameters to specify the settings for an output table. The 'pearsonOut' parameter is specifically for the table containing the requested statistics....
Read full answer See technical action

Yes, you can use the 'inputs' parameter to specify one set of variables and the 'pairWithInput' (or 'with') parameter to specify the second set of variables. The action will then compute correlations for variables between these two sets....
Read full answer See technical action

The countreg.countregFitModel action is used to analyze regression models where the dependent variable consists of non-negative integer or count values. These counts typically represent the number of times an event occurs....
Read full answer See technical action

The action can fit several model types via the 'modeltype' parameter. Supported models include Poisson ('POISSON'), Negative Binomial (type 1 and 2: 'NEGBIN1', 'NEGBIN2'), and Conway-Maxwell-Poisson ('CMPOISSON'). It also handles zero-inflated variations like Zero-Inflated Poisson ('ZIP'), Zero-Infl...
Read full answer See technical action

For data with an excess of zeros, you can use a zero-inflated model by setting the 'modeltype' parameter to 'ZIP', 'ZINB', or 'ZICMPOISSON'. You must also specify the regressors for the zero-inflation component of the model using the 'zeromodel' parameter. The link function for this part can be set ...
Read full answer See technical action
Copyright © WeAreCAS 2023-2025