curvetest (version 2.2)

getoptimalalpha: Test Equality of Curves with Homoscedastic or Heteroscedastic Errors.

Description

To calculate the optiomal smoothing alpha if it si not specified.

Usage

getoptimalalpha(formula, data, plotit = F)

Arguments

formula

Formula to do the regression.

data

A data frame of n rows by 2 columns. The column names should agree with the variable names specified in the formula.

plotit

Whether plot will be geneated to show how the different choices of alpha will affect the generalized cross validation values.

Value

A numeric value of alpha value will be returned.

Details

When this routine is invoked, it will fit a sereies regressions specified by the formula on data set. For each one, the generalized cross validation will be calculated and the "best" (minimal) GCV will be found with the corresponding alpha returned.

References

Zhongfa Zhang, et al: Test Equality of Curves with Homoscedastic or Heteroscedastic Errors. To appear

See Also

curvefit, curvetest.

Examples

    x1=seq(0,1, length=n1<-50); f1<-function(x){x*(1-x)+sin(2*pi*x)}; 
    y1=f1(x1)+rnorm(n1, 0, 0.2) 
    getoptimalalpha(formula=y~x, data.frame(x=x1, y=y1), plotit = TRUE)