Implementation of GMM estimation for Rigobon's (2003) heteroskedasticity-based identification strategy using regime changes.
Usage
rigobon_gmm(
data,
system = "triangular",
y1_var = "Y1",
y2_var = "Y2",
x_vars = "Xk",
regime_var = "regime",
add_intercept = TRUE,
gmm_type = "twoStep",
vcov = .hetid_const("VCOV_HAC"),
initial_values = NULL,
verbose = TRUE,
...
)Arguments
- data
Data frame containing all variables.
- system
Character. Either "triangular" (default) or "simultaneous". Note: Simultaneous systems require many regimes (4+) and large variance differences across regimes for numerical stability and identification.
- y1_var
Character. Name of the first dependent variable.
- y2_var
Character. Name of the second dependent variable.
- x_vars
Character vector. Names of exogenous variables.
- regime_var
Character. Name of the regime indicator variable.
- add_intercept
Logical. Whether to add an intercept.
- gmm_type
Character. GMM type.
- vcov
Character. Variance-covariance matrix type.
- initial_values
Numeric vector. Initial parameter values.
- verbose
Logical. Whether to print progress messages (default: TRUE).
- ...
Additional arguments passed to gmm::gmm.