Munich Personal RePEc Archive

Synthetic Estimation of Dynamic Panel Models When Either N or T or Both Are Not Large: Bias Decomposition in Systematic and Random Components

Carbajal-De-Nova, Carolina and Venegas-Martínez, Francisco (2019): Synthetic Estimation of Dynamic Panel Models When Either N or T or Both Are Not Large: Bias Decomposition in Systematic and Random Components.

[img] PDF
MPRA_paper_94405.pdf

Download (290kB)

Abstract

By increasing the dimensions N or T, or both, in data panel analysis, bias can be reduced asymptotically to zero. This research deals with an econometric methodology to separate and measure bias through synthetic estimators without altering the data panel dimensions. This is done by recursively decomposing bias in systematic and random components. The methodology provides consistent synthetic estimators.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.