utility_by_time_from_death: Calculate utilities based on time before death
In MattWiener/heemodFits: Using Survival Fits with Heemod

Description Usage Arguments Value Examples

Calculate utilities based on time before death

utility_by_time_from_death(x, ...)

## S3 method for class 'cycle_counts'
utility_by_time_from_death(x, util_before_death,
  util_long_before_death, death_state = "Death", discount_rate = 0, ...)

## S3 method for class 'eval_strategy'
utility_by_time_from_death(x, ...)

## S3 method for class 'run_model'
utility_by_time_from_death(x, m, ...)

`x`	an object of of type `cycle_counts`, `eval_model`, or `run_models`. x is either a matrix of counts for from a model. (of type `cycle_counts`) or an `eval_model` or a `run_models` list.
`...`	other arguments to be passed through
`util_before_death`	a matrix with two columns, `through_cycle` and `util`. From `through_cycle[i - 1]` (or 1, for i = 1) to `through_cycle[i+1]` before death, an individual will accrue `util[i]` units of utility.
`util_long_before_death`	utility more than `max(util_before_death[, through_cycle])` cycles before death.
`death_state`	the name of the state representing death.
`discount_rate`	discount rate to be applied to utilities.
`m`	name or number of a model, only if `x` is of type `run_models`.

a vector of utilities, of length nrow(counts).

ProgFree <- round(1000 * exp(-0.4 * 0:24))
Progressive <- round((1000 - ProgFree) * exp(-0.4 * 0:24))
Death <- 1000 - ProgFree - Progressive
state_names <- rep(c("ProgFree", "Progressive", "Death"), each = 25)

counts <- data.frame(.strategy = rep("s1", 25),
                     markov_cycle = 0:24,
                     state_names = state_names,
                     count = c(ProgFree, Progressive, Death)
)
class(counts) <- c("cycle_counts", class(counts))
aa1 <- data.frame(until_lag = c(1,2,3), util = c(0.1, 0.3, 0.5))
 utility_by_time_from_death(counts, util_before_death = aa1, 
   util_long_before_death = 1)