Skip to contents

Rigobon (2003) Regime-Based Heteroskedasticity Identification

Source: R/rigobon.R
run_rigobon_analysis.Rd

This file implements the Rigobon (2003) procedure for using discrete regime indicators to generate instruments for identification in triangular systems with endogenous regressors. The method exploits heteroskedasticity across different regimes (e.g., policy periods, market conditions).

Usage

run_rigobon_analysis(
  n_obs = .hetid_const("N_DEFAULT"),
  params = NULL,
  data = NULL,
  regime_var = "regime",
  endog_var = "Y2",
  exog_vars = "Xk",
  verbose = TRUE,
  return_all = FALSE
)

Arguments

n_obs

Integer. Sample size (default: 1000).

params

List. Parameters for data generation. If NULL, uses default parameters suitable for demonstration.

data

Data.frame. Optional. Pre-existing data with regime indicators. If provided, skips data generation.

regime_var

Character. Name of regime variable in data (default: "regime").

endog_var

Character. Name of endogenous variable (default: "Y2").

exog_vars

Character vector. Names of exogenous variables (default: "Xk").

verbose

Logical. Whether to print progress messages (default: TRUE).

return_all

Logical. Whether to return all intermediate results (default: FALSE).

Value

A list containing:

  • estimates: Data frame comparing OLS and Rigobon 2SLS estimates

  • diagnostics: Heteroskedasticity test results and first-stage F-stats

  • data: The data used (generated or provided)

  • models: Fitted model objects (if return_all = TRUE)

Details

The Rigobon method is a special case of Lewbel's (2012) approach where the heteroskedasticity drivers are discrete regime indicators rather than continuous functions of exogenous variables. Run Complete Rigobon Analysis

This is the main function that performs a complete Rigobon (2003) heteroskedasticity-based identification analysis. It combines data generation, estimation, and diagnostic testing in a single workflow.

References

Rigobon, R. (2003). "Identification Through Heteroskedasticity." The Review of Economics and Statistics, 85(4), 777-792.

Rigobon, R. (2003). Identification through heteroskedasticity. Review of Economics and Statistics, 85(4), 777-792. doi:10.1162/003465303772815727

See also

generate_rigobon_data, run_rigobon_estimation, validate_rigobon_assumptions

Examples

if (FALSE) { # \dontrun{
# Quick analysis with default parameters
results <- run_rigobon_analysis()

# Custom parameters for stronger identification
params <- list(
  beta1_0 = 0.5, beta1_1 = 1.5, gamma1 = -0.8,
  beta2_0 = 1.0, beta2_1 = -1.0,
  alpha1 = -0.5, alpha2 = 1.0,
  regime_probs = c(0.3, 0.7),
  sigma2_regimes = c(1.0, 3.0) # Large difference in variances
)
results <- run_rigobon_analysis(n_obs = 2000, params = params)

# Using existing data
# Assume you have data with a regime indicator
results <- run_rigobon_analysis(data = my_data, regime_var = "period")
} # }