The GofCens package include the following graphical tools and goodness-of-fit tests for complete and right-censored data: - Kolmogorov-Smirnov, Cramér-von Mises, and Anderson-Darling tests, which use the empirical distribution function for complete data and are extended for right-censored data. - Generalized chi-squared-type test, which is based on the squared differences between observed and expected counts using random cells with right-censored data. - A series of graphical tools such as probability or cumulative hazard plots to guide the decision about the most suitable parametric model for the data.
GofCens can be installed from CRAN:
{r CRAN-instalation, eval = FALSE} install.packages("GofCens")
To conduct goodness-of-fit tests with right censored data we can use
the KScens()
, CvMcens()
, ADcens()
and chisqcens()
functions. We illustrate this by means of
the colon
dataset: ```{r, eval = FALSE} #
Kolmogorov-Smirnov set.seed(123) KScens(Surv(time, status) ~ 1, colon,
distr = “norm”)
colonsamp <- colon[sample(nrow(colon), 300), ] CvMcens(Surv(time, status) ~ 1, colonsamp, distr = “normal”)
ADcens(Surv(time, status) ~ 1, colonsamp, distr = “normal”)
chisqcens(Surv(time, status) ~ 1, colonsamp, M = 6, distr = “normal”)
The graphical tools provide nice plots via the functions `cumhazPlot()`, `kmPlot()` and `probPlot()`. See several examples using the `nba` data set:
```{r, eval = FALSE}
data(nba)
cumhazPlot(Surv(survtime, cens) ~ 1, nba, distr = c("expo", "normal", "gumbel"))
kmPlot(Surv(survtime, cens) ~ 1, nba, distr = c("normal", "weibull", "lognormal"),
prnt = FALSE)
probPlot(Surv(survtime, cens) ~ 1, nba, "lognorm", plots = c("PP", "QQ", "SP"),
ggp = TRUE, m = matrix(1:3, nr = 1))