Method of Moments Quantile Regression
Method of Moments for Quantile Regression · Also known as: GMM quantile regression
Method of Moments Quantile Regression combines moment-based estimation (GMM) with quantile regression to estimate distribution parameters while handling endogeneity, panel structure, and dynamic relationships. Introduced by Koenker (2004) and developed by Machado and Mata (2005), it enables distributional analysis (not just mean regression) in complex settings like dynamic panels and instrumental-variable contexts. This approach is powerful for understanding heterogeneity in treatment effects and policy impacts.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use method-of-moments quantile regression when: (1) you need distributional effects (not just means), (2) variables are endogenous, and (3) instrumental variables are available. Valuable for wage inequality studies (returns to education at different wage quantiles despite endogeneity), policy impact heterogeneity, and treatment-effect heterogeneity.
Strengths & limitations
- Handles endogeneity while estimating distributional effects
- Reveals how treatment effects vary across the conditional distribution
- Flexible framework accommodating panel structures and lagged outcomes
- Leverages moment conditions, enabling diverse identification strategies
- Computational complexity; requires careful optimization and numerical stability
- Standard errors can be large in high-dimensional moment sets
- Instrument selection is critical but often ad hoc
- Asymptotics are less developed than for standard quantile or GMM regression alone
Frequently asked
How do I choose instruments in quantile GMM?
Instruments should be uncorrelated with quantile errors and correlated with endogenous variables. Test instrument relevance via first-stage F-statistics; rule of thumb: F > 10. Validity requires economic reasoning.
Should I use many or few moment conditions?
Fewer moments (just-identified) yield lower variance but may be inefficient. More moments (over-identified) improve efficiency but inflate standard errors if misspecified. Start parsimonious; add moments if gains are clear.
How do I estimate standard errors?
Bootstrap is most reliable but computationally intensive. Analytical standard errors exist but require careful implementation. Use software defaults with caution; verify via sensitivity analysis.
Can I compare quantile coefficients across tau?
Yes. Plot coefficient point estimates and confidence bands across quantiles. Test whether slope at 25th percentile differs from 75th percentile using joint tests.
Sources
- Koenker, R. (2004). Quantile regression for longitudinal data. Journal of Multivariate Analysis, 91(1), 74-89. DOI: 10.1016/j.jmva.2004.05.006 ↗
- Machado, J. A., & Mata, J. (2005). Low wage workers and the wage Kuznets curve: Heterogeneity across quantiles. International Journal of Manpower, 26(7-8), 694-712. link ↗
How to cite this page
ScholarGate. (2026, June 3). Method of Moments for Quantile Regression. ScholarGate. https://scholargate.app/en/econometrics/method-of-moments-quantile-regression
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
Compare side by side →