Library: | gam |
See also: | intest gintest gintestpl gamfit pcad |
Quantlet: | intestpl | |
Description: | estimation of the univariate additive functions in a separable additive partial linear model using local polynomial estimation |
Usage: | {m,b,const} = intestpl(x,t,y,h,g,loc{,opt}) | |
Input: | ||
x | n x d matrix, the discrete predictor variables. | |
t | n x p matrix, the continuous predictor variables. | |
y | n x q matrix , the observed response variables | |
h | p x 1 or 1 x 1 matrix , chosen bandwidth for the directions of interest | |
g | p x 1 or 1 x 1 matrix , chosen bandwidth for the directions not of interest | |
loc | dummy , if loc=1 local linear, if loc=2 local quadratic estimator will be used, | |
opt | optional, a list with optional input. The macro "gplmopt" can be used to set up this parameter. The order of the list elements is not important. | |
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). | |
Output: | ||
m | ng x pp x q matrix, where pp = p * loc, containing the marginal integration estimators of the additive components in the first p columns, the derivatives in the following. | |
b | d x 1 matrix, containing the estimators of the linear factors of the corresponding linear influences | |
const | scalar, containing the intercept |
library("gam") randomize(1345) loc= 2 x = matrix(50,2) t = uniform(50,2)*2-1 xh = uniform(50,2) x[,1]= 3*(xh>=0.8)+2*((0.8>xh)&&(xh>=0.3))+(0.3>xh) x[,2]=(xh>(1/3)) g1 = 2*t[,1] g2 =(2*t[,2])^2 g2 = g2 -mean(g2) m = g1 + g2 + x*(0.2|-1.0) y = m + normal(50,1)*0.25 h = #(1.4, 1.4) g = #(1.4, 1.4) {m,b,const} = intestpl(x,t,y,h,g,loc) b const bild =createdisplay(1,2) dat11= t[,1]~g1 dat12= t[,1]~m[,1] setmaskp(dat12,4,4,8) show(bild,1,1,dat11,dat12) dat21= t[,2]~g2 dat22= t[,2]~m[,2] setmaskp(dat22,4,4,8) show(bild,1,2,dat21,dat22)
the marginal integration estimates of the additive functions and the vector of the linear part, see Fan, Haerdle and Mammen (1996) and the derivatives of the additive functions, using local polynomials, see Severance-Lossin & Sperlich (1997).