This could be, for instance, additional information criteria or goodness-of-fit As with the rank argument, each function must accept asĪn argument a fitted model object and return (a value coercible to) a Provided as functions, function names or a list of such (preferably named Optional additional statistics to be included in the result, Otherwise the result may be visually unpleasant. (such as family objects) should be either named (uniquely) or quoted Item names correspond to theĪrguments, and each item provides a list of choices (i.e. Optionally, a named list describing the additional arguments If TRUE or 1, all calls to the fitting functionĪre printed before actual fitting takes place. Logical expression describing models to keep in the resulting Specifying limits as m.min and m.max is allowed for backward See ‘Subsetting’.įor the number of terms in a single model (excluding the intercept). Giving names of terms to be included in all models. Optional, either a single-sided formula or a character vector Optionally, the rank function returning a sort of an informationĬriterion, to be used instead AICc, e.g. Forīackwards compatibility, logical value is also accepted, TRUE is Standardized by SD and Partial SD, respectively.
"none" corresponds to unstandardizedĬoefficients, "sd" and "partial.sd" to coefficients Must be one of "none", "sd" or "partial.sd". Indicates whether and how the coefficients are standardized, and
0 kommentar(er)
