broom (version 0.3.4)

cch_tidiers: tidiers for case-cohort data

Description

Tidiers for case-cohort analyses: summarize each estimated coefficient, or test the overall model.

Usage

## S3 method for class 'cch':
tidy(x, conf.level = 0.95, ...)
## S3 method for class 'cch':
glance(x, ...)

Arguments

x

a "cch" object

conf.level

confidence level for CI

...

extra arguments (not used)

Value

• All tidying methods return a data.frame without rownames, whose structure depends on the method chosen.

  tidy returns a data.frame with one row for each term

• term
  name of term
• estimate
  estimate of coefficient
• stderror
  standard error
• statistic
  Z statistic
• p.value
  p-value
• conf.low
  low end of confidence interval
• conf.high
  high end of confidence interval
• glance returns a one-row data.frame with the following columns:
• score
  score
• rscore
  rscore
• p.value
  p-value from Wald test
• iter
  number of iterations
• n
  number of predictions
• nevent
  number of events

Details

It is not clear what an augment method would look like, so none is provided. Nor is there currently any way to extract the covariance or the residuals.

See Also

cch

Examples

if (require("survival", quietly = TRUE)) {
    # examples come from cch documentation
    subcoh <- nwtco$in.subcohort
    selccoh <- with(nwtco, rel==1|subcoh==1)
    ccoh.data <- nwtco[selccoh,]
    ccoh.data$subcohort <- subcoh[selccoh]
    ## central-lab histology
    ccoh.data$histol <- factor(ccoh.data$histol,labels=c("FH","UH"))
    ## tumour stage
    ccoh.data$stage <- factor(ccoh.data$stage,labels=c("I","II","III" ,"IV"))
    ccoh.data$age <- ccoh.data$age/12 # Age in years

    fit.ccP <- cch(Surv(edrel, rel) ~ stage + histol + age, data = ccoh.data,
                   subcoh = ~subcohort, id= ~seqno, cohort.size = 4028)

    tidy(fit.ccP)

    # coefficient plot
    library(ggplot2)
    ggplot(tidy(fit.ccP), aes(x = estimate, y = term)) + geom_point() +
        geom_errorbarh(aes(xmin = conf.low, xmax = conf.high), height = 0) +
        geom_vline()

    # compare between methods
    library(dplyr)
    fits <- data_frame(method = c("Prentice", "SelfPrentice", "LinYing")) %>%
        group_by(method) %>%
        do(tidy(cch(Surv(edrel, rel) ~ stage + histol + age, data = ccoh.data,
                    subcoh = ~subcohort, id= ~seqno, cohort.size = 4028,
                    method = .$method)))

    # coefficient plots comparing methods
    ggplot(fits, aes(x = estimate, y = term, color = method)) + geom_point() +
        geom_errorbarh(aes(xmin = conf.low, xmax = conf.high)) +
        geom_vline()
}