mdcev.sim: mdcev.sim

In rmdcev: Kuhn-Tucker and Multiple Discrete-Continuous Extreme Value Models

mdcev.sim

Description

Simulate welfare or demand for MDCEV model

Usage

mdcev.sim(
  df_indiv,
  df_common,
  sim_options,
  sim_type = c("welfare", "demand"),
  nerrs = 30,
  cond_error = 1,
  draw_mlhs = 1,
  algo_gen = NULL,
  tol = 1e-20,
  max_loop = 999,
  suppressTime = FALSE,
  stan_seed = 3,
  ...
)

## S3 method for class 'mdcev.sim'
print(
  x,
  digits = max(3, getOption("digits") - 3),
  width = getOption("width"),
  ...
)

## S3 method for class 'mdcev.sim'
summary(object, ci = 0.95, ...)

## S3 method for class 'summary.mdcev.sim'
print(
  x,
  digits = max(3, getOption("digits") - 2),
  width = getOption("width"),
  ...
)

Arguments

df_indiv	Prepared individual level data from PrepareSimulationData
df_common	Prepared common data from PrepareSimulationData
sim_options	Prepared simulation options from PrepareSimulationData
sim_type	Either "welfare" or "demand"
nerrs	Number of error draws for welfare analysis
cond_error	Choose whether to draw errors conditional on actual demand or not. Conditional error draws (=1) or unconditional error draws.
draw_mlhs	Generate draws using Modified Latin Hypercube Sampling algorithm (=1) or uniform (=0)
algo_gen	Type of algorithm for simulation. algo_gen = 0 for Hybrid Approach (i.e. constant alphas, only hybrid models) algo_gen = 1 for General approach (i.e. heterogeneous alpha's, all models)
tol	Tolerance level for simulations if using general approach
max_loop	maximum number of loops for simulations if using general approach
suppressTime	Suppress simulation time calculation
stan_seed	Seed for pseudo-random number generator get_rng see help(get_rng, package = "rstan")
...	Additional parameters to pass to mdcev.sim
x, object	an object of class 'mdcev.sim'
digits	the number of digits,
width	the width of the printing,
ci	choose confidence interval for simulations. Default is 95 percent.

Value

a object of class mdcev.sim which contains a list for each individual holding either 1) nsims x npols matrix of welfare changes if welfare is being simulated or 2) nsims number of lists of npols x # alternatives matrix of Marshallian demands is demand is being simulated.

See Also

[mdcev()] for the estimation of mdcev models.

Examples

data(data_rec, package = "rmdcev")

data_rec <- mdcev.data(data_rec, subset = id <= 500, id.var = "id",
                alt.var = "alt", choice = "quant")

mdcev_est <- mdcev( ~ 0, data = data_rec,
               model = "hybrid0", algorithm = "MLE",
               std_errors = "mvn")

policies <- CreateBlankPolicies(npols = 2,
             mdcev_est,
             price_change_only = TRUE)

df_sim <- PrepareSimulationData(mdcev_est, policies)

wtp <- mdcev.sim(df_sim$df_indiv,
                 df_common = df_sim$df_common,
                 sim_options = df_sim$sim_options,
                 cond_err = 1, nerrs = 5, sim_type = "welfare")

rmdcev documentation built on March 31, 2023, 6:49 p.m.