fetobit {capybara}R Documentation

Tobit model fitting with high-dimensional k-way fixed effects

Description

A wrapper for feglm with family = "tobit". This fits a censored regression model where outcomes below tobit_lb or above tobit_ub are treated as censored. The censoring bounds are specified via the control argument.

Usage

fetobit(
  formula = NULL,
  data = NULL,
  weights = NULL,
  vcov = NULL,
  beta_start = NULL,
  eta_start = NULL,
  offset = NULL,
  tobit_lb = -Inf,
  tobit_ub = Inf,
  control = NULL
)

Arguments

formula an object of class "formula": a symbolic description of the model to be fitted. formula must be of type response ~ slopes | fixed_effects | cluster.
data an object of class "data.frame" containing the variables in the model. The expected input is a dataset with the variables specified in formula and a number of rows at least equal to the number of variables in the model.
weights an optional string with the name of the prior weights variable in data.
vcov an optional character string specifying the type of variance-covariance estimator. One of "iid" (default OLS, ignore cluster part of formula), "hetero" (heteroskedastic-robust HC0, computed in C++ - no cluster variable needed), "cluster" (one-way sandwich using the cluster variable in the formula), "m-estimator" (M-estimator one-way sandwich), or "dyadic" (Cameron-Miller dyadic sandwich; requires two entity variables in the third part of the formula). When NULL (default), the type is inferred from the formula: if a cluster variable is present the standard sandwich is used, otherwise the inverse Hessian (IID) is returned.
beta_start an optional vector of starting values for the structural parameters in the linear predictor. Default is \boldsymbol{\beta} = \mathbf{0}.
eta_start an optional vector of starting values for the linear predictor.
offset an optional formula or numeric vector specifying an a priori known component to be included in the linear predictor. If a formula, it should be of the form ~ variable.
tobit_lb numeric indicating the lower censoring bound. Observations with y <= tobit_lb are treated as left-censored. Default is -Inf (no left censoring).
tobit_ub numeric indicating the upper censoring bound. Observations with y >= tobit_ub are treated as right-censored. Default is Inf (no right censoring).
control a named list of parameters for controlling the fitting process. See fit_control for details.

Value

A named list of class "feglm".

See Also

fit_control for additional control parameters

Examples

# Left-censored at 0 (Type I Tobit)
d <- data.frame(
  y = pmax(0, rnorm(100, 2, 1)),
  x = rnorm(100),
  g = factor(rep(1:5, 20))
)
mod <- fetobit(y ~ x | g, d, tobit_lb = 0)
summary(mod)

# Two-sided censoring
d2 <- data.frame(
  y = pmin(pmax(rnorm(100, 5, 2), 0), 10),
  x = rnorm(100),
  g = factor(rep(1:5, 20))
)
mod2 <- fetobit(y ~ x | g, d2, tobit_lb = 0, tobit_ub = 10)
summary(mod2)

Loading...