# aj: aj In mbrueckner/AHR: Estimation and Testing of Average Hazard Ratios

## Description

Aalen-Johansen estimator (empirical transition matrix)

## Usage

 `1` ```aj(times, data, param) ```

## Arguments

 `times` a vector of evaluation times `data` data frame (see `etm` function documentation) `param` list of parameters (target, states, transitions, censoring, s, t, cov) (see etm documentation)

## Details

Wrapper for 'etm' function from the 'etm' package to be used with `ahrAJ`

For a description of the parameters in the list `param` see the documentation of the `etm` function in package `etm`.

## Value

a list containing

 `times` the argument `times` passed to the function `S` vector of 1 minus transition probabilities at `times` (one for each element of `times`) `V` vector of variances at 'times' (only if `param\$cov` is TRUE) `logCOV` matrix containing estimated values of the log-covariance function evaluated for all pairs of elements of the vector `times` (only if `param\$cov` is TRUE and model is a competing risks model)

`etm`
 ``` 1 2 3 4 5 6 7 8 9 10 11 12``` ```## competing risks T <- rexp(100) C <- rexp(100) r <- rbinom(100, 2, 0.5) r[(r == 0) | (T > C)] <- "cens" data <- data.frame(id=1:100, time=pmin(T,C), from=rep(0, 100), to=r) data <- data[order(data\$time),] tra <- matrix(FALSE, nrow=3, ncol=3) tra[1, 2:3] <- TRUE # estimate cumulative incidence function for event type 1 fit <- aj(sort(data\$time), data, list(target="0 1", states=c("0", "1", "2"), transitions=tra, censoring="cens", s=0, t="last", covariance=TRUE)) ```