Learn R Programming

OutlierDC (version 0.3-0)

OutlierDC-package: Functions for detecting outlying observations for censored data using quantile regression

Description

This package offers three outlier detection algorithms for censored data using quantile regression.

Arguments

Details

Package:
OutlierDC
Type:
Package
Version:
0.3-0
Date:
2014-03-23
License:
GPL version 3
LazyLoad:
no

References

Eo S-H, Hong S-M Hong, Cho H (2014). Identification of outlying observations with quantile regression for censored data, Submitted.

Wang HJ, Wang L (2009) Locally Weighted Censored Quantile Regression. JASA 104:1117--1128. doi: 10.1198/jasa.2009.tm08230

See Also

odc, plot, coef, show, quantreg

Examples

Run this code
  ## Not run: 
#     require(OutlierDC)
#     # Toy example 
#     data(ebd)
#     # The data consists of 402 observations with 6 variables. 
#     dim(ebd)
#     # To show the first six observations of the dataset,
#     head(ebd)
#     
#     #scoring algorithm
#     fit <- odc(Surv(log(time), status) ~ meta, data = ebd)
#     fit
#     coef(fit)
#     plot(fit)
# 
#     # Add upper bound for the selection of outleirs
#     fit1 <- update(fit, k_s = 4)
#     fit1
#     plot(fit1)
# 
#     # residual-based algorithm
#     fit2 <- odc(Surv(log(time), status) ~ meta, data = ebd, method = "residual", k_r = 1.5)
#     fit2
#     plot(fit2)
#     
#     # To display all of outlying observations in the fitted object
#     fit2@outlier.data
#     
#     # boxplot algorithm
#     fit3 <- odc(Surv(log(time), status) ~ meta, data = ebd, method = "boxplot", k_b = 1.5)
#     fit3
#     plot(fit3, ylab = "log survival times", xlab = "metastasis lymph nodes")
# ## End(Not run)

Run the code above in your browser using DataLab