stratEst.simulate function

Simulation function for strategy estimation.

The simulation function of the package.

stratEst.simulate(
  data = NULL,
  strategies,
  shares = NULL,
  coefficients = NULL,
  covariate.mat = NULL,
  num.ids = 100,
  num.games = 5,
  num.periods = NULL,
  fixed.assignment = TRUE,
  input.na = FALSE,
  sample.id = NULL
)

Arguments

• data: a stratEst.data object. Alternatively, the arguments num.ids, num.games, and num.periods can be used if no data is available.
• strategies: a list of strategies. Each element if the list must be an object of class stratEst.strategy.
• shares: a numeric vector of strategy shares. The order of the elements corresponds to the order in strategies. NA values are not allowed. Use a list of numeric vectors if data has more than one sample and shares are sample specific.
• coefficients: a matrix of regression coefficients. Column names correspond to the names of the strategies, row names to the names of the covariates.
• covariate.mat: a matrix with the covariates in columns. The column names of the matrix indicate the names of the covariates. The matrix must have as many rows as there are individuals.
• num.ids: an integer. The number of individuals. Default is 100.
• num.games: an integer. The number of games. Default is 5.
• num.periods: a vector of integers with as many elements num.games. The elements specify the number of periods in each game. Default (NULL) means 5 periods in each game.
• fixed.assignment: a logical value. If FALSE individuals use potentially different strategies in each each game. If TRUE, individuals use the same strategy in each game. Default is FALSE.
• input.na: a logical value. If FALSE an input value is randomly selected for the first period. Default is FALSE.
• sample.id: a character string indicating the name of the variable which identifies the samples in data. Individual observations must be nested in samples. Default is NULL.

Returns

A stratEst.data object. A data frame in the long format with the following variables: - id: the variable that identifies observations of the same individual.

• game: the variable that identifies observations of the same game.

• period: the period of the game.

• choice: the discrete choices.

• input: the inputs.

• sample: the sample of the individual.

• strategy: the strategy of the individual.

Examples

## Simulate data of two strategies for choices "left" and "right".
lr <- c("left","right")
pi <- runif(1)
pr <- c(1,0,0,1)
tr <- c(1,2,1,2)
mixed <- stratEst.strategy(choices = lr, inputs = lr, prob.choices = c(pi, 1 - pi))
pure <- stratEst.strategy(choices = lr, inputs = lr, prob.choices = pr, tr.inputs = tr)
gamma <- runif(1)/4
pure$tremble <- gamma
beta <- rnorm(1)
p <- 1/sum(1 + exp(beta))
sim.shares <- c(p, 1-p)
sim.strategies <- list("mixed" = mixed, "pure" = pure)
sim.data <- stratEst.simulate(strategies = sim.strategies, shares = sim.shares)