gplmbilo | GPLM (logit) -- gplmbilo fits a generalized partially linear model where y|x,t is binomial distributed and E[y|x,t] and x*b + m(t) are linked via the logistic function (canonical link). |
gplmbilobiased | biased GLM (logit) -- gplmbilobiased computes the biased generalized linear model for the test of a GLM (logit) versus a GPLM (logit). |
gplmbilobootstraptest | Bootstrap test (using as. normality) GLM (logit) vs. GPLM (logit) -- by default gplmnoidtest tests the generalized "truly" linear model E[y|x,t] = G(x*b + t*g + c) versus the genralized partially linear model E[y|x,t] = G(x*b + m(t)), where G is the logistic link function. Optional, an alternative design matrix can be specified for t or parametric estimates for b and m can be given directly. |
gplmbilotest | Test (using as. normality) GLM (logit) vs. GPLM (logit) -- by default gplmnoidtest tests the generalized "truly" linear model E[y|x,t] = G(x*b + t*g + c) versus the genralized partially linear model E[y|x,t] = G(x*b + m(t)), where G is the logistic link function. Optional, an alternative design matrix can be specified for t or parametric estimates for b and m can be given directly. |
gplmcore | gplmcore fits a generalized partially linear model E(y|x,t) = G(x*b + m(t)). This is the core macro for GPLM estimation. It assumes that all input variables are given in the right manner. No preparation of data is performed. A more convenient way to estimate a GPLM is to call the function gplmest. |
gplmest | gplmest fits a generalized partially linear model E[y|x,t] = G(x*b + m(t)). This macro offers a convenient interface for GPLM estimation. A preparation of data is performed (inclusive sorting). |
gplminit | gplminit checks the validity of input and performs the initial calculations for an GPLM fit (inclusive sorting). The output is ready to be used with gplmcore. |
gplmmain | loads everything necessary for library gplm. |
gplmnoid | PLM -- gplmnoid fits a generalized partially linear model where y|x,t is normally distributed and E[y|x,t] = x*b + m(t) (canonical link). |
gplmnoidbiased | biased LM -- gplmnoidbiased computes the biased linear model for the test of a linear model versus a PLM. This is a fast routine using the command sker to obtain kernel estimates. |
gplmnoidtest | Test (using as. normality) linear vs. partially linear -- by default gplmnoidtest tests the "truly" linear model E[y|x,t] = x*b + t*g + c versus the partially linear model E[y|x,t] = x*b + m(t). Optional, an alternative design matrix can be specified for t or parametric estimates for b and m can be given directly. |
gplmopt | gplmopt defines a list with optional parameters in gplm macros. The list is either created or new options are appended to an existing list. Note that gplmopt does accept _any_ values for the parameters without validity. |
gplmout | gplmout creates a nice output display for gplm. |
gplmstat | gplmstat provides some statistics for a fitted GPLM. |
gplmtest | gplmtest verifies the GPLM macros. |
replicdata | replicdata reduces a matrix x to its distinct rows and gives the number of replications of each rows in the original dataset. An optional second matrix y can be given, the rows of y are summed up accordingly. replicdata does in fact the same as discrete but provides an additional index vector to identify the reduced data with the original. It takes a little longer due an additonal sort. |