Econometric studies on flexible modeling of developing countries in growth analysis

Ökonometrische Studien über Wachstumsanalysen von Entwicklungsländern

by Max Köhler
Doctoral thesis
Date of Examination:2012-05-02
Date of issue:2012-05-25
Advisor:Prof. Dr. Stefan Sperlich
Referee:Prof. Dr. Stefan Sperlich
Referee:Prof. Dr. Inmaculada Martínez-Zarzoso
Referee:Prof. Dr. Thomas Kneib
Persistent Address: http://dx.doi.org/10.53846/goediss-3012

 

 

Files in this item

Name:koehler.pdf
Size:1.58Mb
Format:PDF

The following license files are associated with this item:

Abstract

English

Over the last four decades, several methods for selecting the smoothing parameter, generally called the bandwidth, have been introduced in kernel regression. They differ quite a bit, and although there already exist more selection methods than for any other regression smoother we can still see coming up new ones. Given the need of automatic data-driven bandwidth selectors for applied statistics, this review is intended to explain and compare these methods. The Africa-Dummy has been identified and different explanations for its appearance have been published. In this paper, the issue of the empirical identification of the Africa-Dummy is addressed. We introduce a fixed effects regression model to identify the Africa-Dummy in one regression step so that its correlations to other coefficients can be estimated. A semiparametric extension of this model checks whether the Africa-Dummy is a result of misspecification of the functional structure. Furthermore, we show that sub-Saharan African countries have a positive return to the population growth and when adding interaction effects, the Africa-Dummy is even positive. Moreover, we show that the Africa-Dummy changes dramatically over time and the punishment for sub-Saharan African countries decreases incrementally since the mid-nineties. According to the Augmented Solow Growth model, it was even insignificant since the end-nineties. Various papers demonstrate the importance of inequality, poverty and the size of the middle class for economic growth. When explaining why these measures of the income distribution are added to the growth regression, it is often mentioned that poor people behave different than rich people which translates to the economy as a whole. However, adding explanatory variables does not reflect this behavior. We formulate and apply a variable-coefficients model and show that the coefficients of the growth regressions differ a lot and this can be explained by the level of poverty, inequality and the middle class. Furthermore, we investigate how the coefficients and therefore the growth path differs for the poorer and for the richer part of the society. We argue that the differences in the coefficients impeach, on the one hand, the credibility of that the mean coefficients are informative, and, on the other hand, the credibility of the economic justification for explaining the growth path of a country with mean coefficients. Moreover, we explain that, when estimating mean coefficients, the estimation is likely to suffer from an endogeneity and from a sample selection bias.
Keywords: Nonparametric Regression; Economic Growth; Panel Data
Schlagwörter: Nichtparametrische Regression; Ökonomisches Wachstum; Paneldaten
 
Statistik