 Usage:  gamfit(t,y{,opt})  
 
 Input:

  t                      n x p matrix, the predictor variables. 
                         
  y                      n x 1 vector , the observed 
                         response variables 
                         
  opt                    optional, a list with optional input. The macro 
                         "gamopt" 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 
                         deflts (see below). 
                         
  opt.x                  n x d matrix, the discrete predictor variables. 
                         
  opt.h                  p x 1 or 1 x 1 matrix , chosen bandwidth for 
                         the directions of interest 
                         
  opt.g                  p x 1 or 1 x 1 matrix , chosen bandwidth for 
                         the directions not of interest 
                         
  opt.loc                dummy , for loc=0 local constant (Nad. Wats.), 
                         for loc=1 local linear and for loc=2 local 
                         quadratic estimator will be used 
                         
  opt.kern               string, specifying the kernel function for backfitting 
                         algorithm 
                         
  opt.tg                 ng x pg vector, a grid for continuous part. If tg is 
                         given, the nonparametric function will be 
                         computed on this grid. 
                         
  opt.shf                integer, (show-how-far) if exists and =1, an output 
                         is produced which indicates how the iteration 
                         is going on (additive function / point of estimation / 
                         number of iteration). 
                         
  opt.code               text string, the short code for the model (e.g. 
                         "bilo" for logit or "noid" for ordinary PLM). 
                         
  opt.wx                 scalar or n x 1 vector, prior weights. If not given, 
                         all weights are 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, "gam" 
                         is used. 
                         
  opt.xvars              p x 1 string vector, variable names for the output. 
                         Note, that only up to 11 characters are used. 
                         
 Output:

  gampic or opt.name+"pic"  display, containing estimation result 
                         
  gamfit                 list object, which is made 
                         globally available 
                         
  gamfit.m               estimator for the non-linear part 
                         
  gamfit.b               p x 1 vector, estimated coefficients of the linear part 
                         
  gamfit.bv              p x p matrix, estimated covariance matrix for b. 
                         
  gamfit.const           scalar, estimated constant. 
                         
  gamfit.opt             list, internally used option list. 
                         
--------------------------------------------------------------
(C) MD*TECH Method and Data Technologies, 17.8.2000
