Module Reference

Instrumental Variable Estimation

IV2SLS(dependent, exog, endog, instruments, *)

Estimation of IV models using two-stage least squares.

IVLIML(dependent, exog, endog, instruments, *)

Limited information ML and k-class estimation of IV models.

IVGMM(dependent, exog, endog, instruments, *)

Estimation of IV models using the generalized method of moments (GMM).

IVGMMCUE(dependent, exog, endog, instruments, *)

Estimation of IV models using continuously updating GMM.

IVJIVE(dependent, exog, endog, instruments, *)

Jackknife instrumental variables estimator (JIVE).

_OLS(dependent, exog, *[, weights])

Compute OLS estimates when required.

Absorbing Least Squares

OLS and WLS with high-dimensional effects.

AbsorbingLS(dependent[, exog, absorb, ...])

Linear regression with high-dimensional effects.

AbsorbingLSResults(results, model)

Results from the estimation of a model with absorbed effects.

Interaction([cat, cont, nobs])

Class that simplifies specifying interactions.

AbsorbingRegressor(*[, cat, cont, ...])

Construct a sparse matrix of regressors from components.

Estimation Results

IVResults(results, model)

Results from IV estimation.

IVGMMResults(results, model)

Results from GMM estimation of IV models.

OLSResults(results, model)

Results from OLS model estimation.

IVModelComparison(results, *[, precision, stars])

Comparison of multiple models.

FirstStageResults(dep, exog, endog, instr, ...)

First stage estimation results and diagnostics.

compare(results, *[, precision, stars])

Compare the results of multiple models.

Tests of the First-Stage Rank

cragg_donald(endog, instr, exog)

Cragg-Donald test of reduced rank for the first-stage regression.

cragg_donald_f(endog, instr, exog)

Cragg-Donald Wald F statistic.

kleibergen_paap(endog, instr, exog[, ...])

Kleibergen-Paap rk LM test of underidentification.

kleibergen_paap_f(endog, instr, exog[, ...])

Kleibergen-Paap rk Wald F statistic.

Instrumental Variable Covariance Estimation

HomoskedasticCovariance(x, y, z, params[, ...])

Covariance estimation for homoskedastic data.

HeteroskedasticCovariance(x, y, z, params[, ...])

Covariance estimation for heteroskedastic data.

ClusteredCovariance(x, y, z, params[, ...])

Covariance estimation for clustered data.

KernelCovariance(x, y, z, params[, kernel, ...])

Kernel weighted (HAC) covariance estimation.

Kernel Weight Generators

kernel_weight_bartlett(bw, *args)

Kernel weights from a Bartlett kernel.

kernel_weight_parzen(bw, *args)

Kernel weights from a Parzen kernel.

kernel_weight_quadratic_spectral(bw, n)

Kernel weights from a quadratic-spectral kernel.

kernel_optimal_bandwidth(x[, kernel])

Select the bandwidth of a kernel covariance estimator from the data.

GMM Weight and Covariance Estimation

IVGMMCovariance(x, y, z, params, w[, ...])

Covariance estimation for GMM models.

HomoskedasticWeightMatrix([center, debiased])

Homoskedastic (unadjusted) weight estimation.

HeteroskedasticWeightMatrix([center, debiased])

Heteroskedasticity robust weight estimation.

KernelWeightMatrix([kernel, bandwidth, ...])

Heteroskedasticity, autocorrelation robust weight estimation.

OneWayClusteredWeightMatrix(clusters[, ...])

Clustered (one-way) weight estimation.

IV Data Structures

IVData(x[, var_name, nobs, convert_dummies, ...])

Type abstraction for use in univariate models.