Model-diagnostics based on cumulative residuals

Model-diagnostics based on cumulative residuals

References

D.Y. Lin and L.J. Wei and Z. Ying (2002) Model-Checking Techniques Based on Cumulative Residuals. Biometrics, Volume 58, pp 1-12.

John Q. Su and L.J. Wei (1991) A lack-of-fit test for the mean function in a generalized linear model. Journal. Amer. Statist. Assoc., Volume 86, Number 414, pp 420-426.

See also

cox.aalen in the timereg-package for similar GoF-methods for survival-data.

Author

Klaus K. Holst <klaus@holst.it>

Examples

example(cumres)
#> 
#> cumres> sim1 <- function(n=100, f=function(x1,x2) {10+x1+x2^2}, sd=1, seed=1) {
#> cumres+   if (!is.null(seed))
#> cumres+     set.seed(seed)
#> cumres+   x1 <- rnorm(n);
#> cumres+   x2 <- rnorm(n)
#> cumres+   X <- cbind(1,x1,x2)
#> cumres+   y <- f(x1,x2) + rnorm(n,sd=sd)
#> cumres+   d <- data.frame(y,x1,x2)
#> cumres+   return(d)
#> cumres+ }
#> 
#> cumres> d <- sim1(100); l <- lm(y ~ x1 + x2,d)
#> 
#> cumres> system.time(g <- cumres(l, R=100, plots=50))
#>    user  system elapsed 
#>   0.018   0.000   0.018 
#> 
#> cumres> g
#> 
#>           p-value(Sup) p-value(L2)
#> predicted         0.26        0.37
#> x1                0.47        0.42
#> x2                0.00        0.00
#> 
#> Based on 100 realizations.
#> 
#> cumres> ## No test: 
#> cumres> ##D plot(g)
#> cumres> ## End(No test)
#> cumres> g1 <- cumres(l, c("y"), R=100, plots=50)
#> 
#> cumres> g1
#> 
#>           p-value(Sup) p-value(L2)
#> predicted         0.35        0.35
#> 
#> Based on 100 realizations.
#> 
#> cumres> g2 <- cumres(l, c("y"), R=100, plots=50, b=0.5)
#> 
#> cumres> g2
#> 
#>           p-value(Sup) p-value(L2)
#> predicted         0.21        0.26
#> 
#> Based on 100 realizations.