An Integrated Panel Data Approach to Modelling Economic Growth

Abstract

Empirical growth analysis has three major problems -variable selection, parameter heterogeneity and cross-sectional dependence -which are addressed independently from each other in most studies. The purpose of this study is to propose an integrated framework that extends the conventional linear growth regression model to allow for parameter heterogeneity and cross-sectional error dependence, while simultaneously performing variable selection. We also derive the asymptotic properties of the estimator under both low and high dimensions, and further investigate the finite sample performance of the estimator through Monte Carlo simulations. We apply the framework to a dataset of 89 countries over the period from 1960 to 2014. Our results reveal some cross-country patterns not found in previous studies (e.g., "middle income trap hypothesis", "natural resources curse hypothesis", "religion works via belief, not practice", etc.).

Enhancing Sustainable and Inclusive Growth in the Central African Economic and Monetary Community

“…We choose the initial level of real GDP per capita as a conditional variable. This choice is consistent with the existing growth literature and guided by the optimal level of development variable selected by Feng and others (2019). Because our research focuses on the growth dynamics of an oil-producing region, we also examine the results of our semiparametric estimations by splitting our initial sample between oil and non-oil economies.…”

“…Growth regression models conventionally face three types of problem: variable selection, cross-sectional dependence, and parameter heterogeneity (Feng and others 2019). Variable selection arises when numerous variables can explain a country's growth performance, and these candidates vary across economies.…”

“…We account for this issue by taking three-year averages to mitigate the impact of common shocks on our estimates. approach recently explored in the literature and subject to future research would be to introduce a vector of unobservable common factors that captures country-specific responses to common shocks (Feng and others 2019).…”

“…This may weaken the confidence that policymakers place on empirical results (Hineline 2008). To account for parameter heterogeneity, researchers have developed semiparametric estimation models that allow coefficients to vary across observations according to specified initial conditions (Feng and others 2019;Durlauf, Kourtellos, and Minkin 2001). We revisit the linear assumption of our earlier growth model (equation 6) by allowing our set of control variables to vary according to a country's initial state of development.…”