 Usage:  opt = gamopt(s0,v0{,s1,v1{,...{,opt0}}})  
 
 Input:

  s0,s1,...              string, name of the component to add. Allowed are: 
                         "x" (the discrete predictor variables) 
                         "loc" (dummy for local const, linear or quadr. estim.) 
                         "tg" (grid) 
                         "h" (bandwidth for the directions of interest) 
                         "g" (bandwidth for the directions not of interest) 
                         "code" (model code), 
                         "kern" (kernel function), 
                         "wx" (weights), 
                         "off" (offset), 
                         "shf" (show iteration), 
                         "miter" (maximal number of iterations), 
                         "cnv" (convergence criterion), 
                         "fscor" (Fisher scoring), 
                         "pow" (power), 
                         "nopic" (no picture), 
                         "descript" (add descriptive statistics), 
                         "pl" (partially linear model), 
                         "name" (output name), 
                         "xvars" (discrete variable names), 
                         "tvars" (variable names), 
                         "yvars" (name of y-variable), 
                         "title" (output title), 
                         "bv" (covariance matrix for b). 
                         
  v1,v2,...              object, value of the corresponding component to add. 
                         
  opt0                   optional, name of list where the component(s) 
                         should be added. 
                         
 Output:

  opt                    resulting list of options. 
                         
--------------------------------------------------------------
(C) MD*TECH Method and Data Technologies, 17.8.2000
