Seminar: Uniform inference in linear panel data models with two-dimensional heterogeneity
The Department of Econometrics and Business Statistics invites you to a research seminar 'Uniform inference in linear panel data models with two-dimensional heterogeneity' presented by Professor Liangjun Su from Singapore Management University.
All welcome, no registrations required. Monash staff can register their interest on Workplace.
Abstract
This paper studies uniform inference in linear panel data models when the slope coefficients may exhibit heterogeneity over both the time and individual dimensions and conventional fixed effects estimators may fail to be consistent. We decompose the slope coefficient vector into a common component, an individual heterogeneity component and a time heterogeneity component.
Due to the fact that one can not eliminate either the individual or time heterogeneity in the slope coefficients through the usual demeaning or differencing procedure and that the separate estimates of the individual and time heterogeneity components via respective time series and cross-sectional regressions are generally inconsistent when these components are allowed to be correlated with the regressors, we propose to generalise the fixed effects estimator to estimate these three components along with the traditional individual and time fixed effects in the models under suitable identification restrictions.
To establish the asymptotic properties of the generalised fixed effects (GFE) estimators, we invert a number of large dimensional square matrices by approximating them with quasi-Kronecker structured matrices.
We establish the asymptotic normality of our GFE estimators by characterising their asymptotic biases and variances. The convergence rates of our estimators depend on the unknown degrees of individual and time heterogeneity. To make a uniform inference on the common slope component, we propose a novel triple-bootstrap procedure to estimate the asymptotic variance of our GFE estimators which are valid uniformly over a broad space of parameter heterogeneity. Simulations show the superb performance of our estimators in various scenarios.
We apply our method to study the relationship between savings and investments in a cross-countries study. We find significant parameter heterogeneity along both the individual and time dimensions and provide some new insight to explain the Feldstein-Horioka puzzle.
Event Details
- Date:
- 2 November 2018 at 2:00 pm – 3:30 pm
- Venue:
- Room H4.87, building H, Caulfield campus, 900 Dandenong Road, Caulfield East
- Categories:
- Econometrics and Business Statistics
Description
The Department of Econometrics and Business Statistics invites you to a research seminar 'Uniform inference in linear panel data models with two-dimensional heterogeneity' presented by Professor Liangjun Su from Singapore Management University.
All welcome, no registrations required. Monash staff can register their interest on Workplace.
Abstract
This paper studies uniform inference in linear panel data models when the slope coefficients may exhibit heterogeneity over both the time and individual dimensions and conventional fixed effects estimators may fail to be consistent. We decompose the slope coefficient vector into a common component, an individual heterogeneity component and a time heterogeneity component.
Due to the fact that one can not eliminate either the individual or time heterogeneity in the slope coefficients through the usual demeaning or differencing procedure and that the separate estimates of the individual and time heterogeneity components via respective time series and cross-sectional regressions are generally inconsistent when these components are allowed to be correlated with the regressors, we propose to generalise the fixed effects estimator to estimate these three components along with the traditional individual and time fixed effects in the models under suitable identification restrictions.
To establish the asymptotic properties of the generalised fixed effects (GFE) estimators, we invert a number of large dimensional square matrices by approximating them with quasi-Kronecker structured matrices.
We establish the asymptotic normality of our GFE estimators by characterising their asymptotic biases and variances. The convergence rates of our estimators depend on the unknown degrees of individual and time heterogeneity. To make a uniform inference on the common slope component, we propose a novel triple-bootstrap procedure to estimate the asymptotic variance of our GFE estimators which are valid uniformly over a broad space of parameter heterogeneity. Simulations show the superb performance of our estimators in various scenarios.
We apply our method to study the relationship between savings and investments in a cross-countries study. We find significant parameter heterogeneity along both the individual and time dimensions and provide some new insight to explain the Feldstein-Horioka puzzle.
Event Contact
- Name
- buseco-events@monash.edu
- Phone
- Organisation