nll_controls: Negative log-likelihood function: control data only

Description

Function returning negative log-likelihood (nll) for data in a control treatment.

Usage

nll_controls(
  a1 = a1,
  b1 = b1,
  data = data,
  time = time,
  censor = censor,
  d1 = "Weibull"
)

Arguments

a1, b1

location and scale parameters for background mortality

data

name of data frame containing survival data

time

name of data frame column identifying time of event; time > 0

censor

name of data frame column idenifying if event was death (0) or right-censoring (1)

name of probability distribution describing background mortality. Choice of; 'Weibull', 'Gumbel', 'Fr<U+00E9>chet'; defaults to the Weibull distribution

Value

numeric

Details

This function returns the nll based on two parameters, the location and scale parameters used to describe background mortality.

Examples

Run this code

# NOT RUN {
# prepare a subset of the Blanford data for analysis
  data01 <- subset(data_blanford,
    (data_blanford$block == 3) &
      (data_blanford$treatment == 'cont') &
        (data_blanford$day > 0))

# check data frame for names of columns
    head(data01)

# step #1: 'prep function' linking 'nll_controls' to data
  # and identifying parameters to estimate
    m01_prep_function <- function(a1 = a1, b1 = b1){
      nll_controls(
        a1 = a1, b1 = b1,
        data = data01,
        time = t,
        censor = censor,
        d1 = 'Weibull'
        )}

# step #2: send 'prep_function' to mle2 for maximum likelihood estimation
  # specifying starting values
    m01 <- mle2(m01_prep_function,
             start = list(a1 = 2, b1 = 0.5)
             )

    summary(m01)


# }