 Usage:  doglm(x,y{,opt})  

 

 Input:



  x                      n x p matrix, the predictor variables. 

                         

  y                      n x 1 vector, the response variables. 

                         (In the case of replicated data, the number of 

                         replications should be given in opt.wx and y should 

                         contain the sums of all responses for a replication.) 

                         

  opt                    optional, a list with optional input. The macro 

                         "glmopt" can be used to set up this parameter. 

                         The order of the list elements is not important. 

                         Parameters which are not given are replaced by 

                         defaults (see below). 

                         

  opt.code               text string, the short code for the model (e.g. 

                         "bilo" for logit or "noid" for ordinary PLM). 

                         

  opt.weights            string, type of weights. Can be "frequency" 

                         for replication counts, or "prior" (default) 

                         for prior weights in weighted regression. 

                         

  opt.wx                 scalar or n x 1 vector, frequency or prior 

                         weights. If not given, set to 1. 

                         

  opt.off                scalar or n x 1 vector, offset in linear predictor. 

                         If not given, set to 0. 

                         

  opt.name               string, prefix for the output. If not given, "glm" 

                         is used. 

                         

  opt.xvars              p x 1 string vector, variable names for the output. 

                         Note, that only up to 15 characters are used. 

                         

  opt.title              string, title for the output. If not given, a default 

                         is set. 

                         

  opt.nopic              integer, if exists and =1, output is not shown 

                         in displays, but written to XploRe_out. 

                         

 Output:



  doglm                  or opt.name, 

                         list, global variable with the components: 

                         

  doglm.b                p x 1 vector, estimated coefficients. 

                         

  doglm.bv               p x p matrix, estimated covariance matrix for b. 

                         

  doglm.stat             list with components as computed by glmstat: 

                         serror (standard errors of coefficients), 

                         tvalue (t-values for coefficients), 

                         pvalue (p-values for coefficients), 

                         df (degrees of freedom), 

                         deviance (deviance), 

                         pearson (generalized pearson's chi^2), 

                         loglik (log-likelihood), 

                         dispersion (estimated dispersion =pearson/df), 

                         r2 ((pseudo) coefficient of determination), 

                         adr2 (adjusted (pseudo) coefficient of determination), 

                         aic (Akaike's AIC criterion), 

                         bic (Schwarz' BIC criterion), and 

                         it (number of iterations needed), 

                         ret (return code, 

                         0 if everything went o.k., 

                         1 if maximal number of iterations reached, 

                         -1 if missing values have been encountered), 

                         nr (number of replications found in x). 

                         

  doglmSummary           or opt.name+"Summary", 

                         display, containing summary statistics and 

                         show which variables are selected for fit. 

                         

  doglmsum               or opt.name+"sum", 

                         string vector, global, containing summary 

                         statistics as text. 

                         

  doglmOutput            or opt.name+"Output", 

                         display, containing estimation result in the left 

                         panel and a plot of the index x*b{+off} vs. y 

                         (vs. y./wx for binomials) and a plot of the index 

                         vs. the link function. 

                         

  doglmout               or opt.name+"out", 

                         string vector, global, containing estimation result 

                         in the left panel as text. 

                         

--------------------------------------------------------------

(C) MD*TECH Method and Data Technologies, 21.9.2000

