|
|
Meta-Analysis in Economics: Application to Measuring the Euro's Trade Effect
Polák, Petr ; Havránek, Tomáš (advisor) ; Havránková, Zuzana (referee)
Meta-analysis is a very strong and effective tool designed for the synthesis of results of empirical research. It provides a possibility to make reliable conclusions and offers more systematic and unbiased view at empirical studies than do narrative reviews. This thesis begins with description of meta-analysis from the theoretical point of view and, therefore, is the first Czech-written methodology of modern meta-analysis suitable for economics. This part is followed by an applied meta-analysis that investigates the euro effect on common trade exchange, and the analysis is focused on publication bias and the use of the multilevel random effects model. The empirical part is based on 2580 estimates gathered from 33 studies that investigate the relationship between euro and trade volume. The meta-analysis reveals the presence of publication bias, confirms the economic research cycle hypothesis and estimates, according to the available literature, that the true Rose effect lies probably between 2 and 6 percent.
|
|
|
Euro and the Effect on Bilateral Trade: Gravity Model Analysis
Gabaš, Ondřej ; Semerák, Vilém (advisor) ; Paulus, Michal (referee)
Euro and the Effect on Bilateral Trade: Gravity Model Analysis Ondřej Gabaš The purpose of this thesis is to estimate the impact of common Euro- pean currency on bilateral trade. Using data from 1993 to 2015, we employ structural Gravity model of trade on the sample of 19 European Monetary Union members and 25 developed countries. In our analysis, we use two different methods to account for the endogeneity, country-pair fixed-effects and Baier and Bergstrands (2009) specification of the Gravity model esti- mated by OLS. In order to examine the effect of missing observations, we employ Poisson Pseudo-Maximum-Likelihood estimator. Last, we focus on the adoption of the Euro as a reason behind the effect of trade diversion. The results of all three models show that the creation of the EMU had sta- tistically insignificant effect limiting to zero. In the case of fixed-effects and OLS, the estimated effect is negative, while in the case of the PPML, we found a positive impact. In addition, the results of our analysis show that the adoption of the Euro did not cause a trade diversion. 1
|
|
|
Meta-Analysis in Economics: Application to Measuring the Euro's Trade Effect
Polák, Petr ; Havránek, Tomáš (advisor) ; Havránková, Zuzana (referee)
Meta-analysis is a very strong and effective tool designed for the synthesis of results of empirical research. It provides a possibility to make reliable conclusions and offers more systematic and unbiased view at empirical studies than do narrative reviews. This thesis begins with description of meta-analysis from the theoretical point of view and, therefore, is the first Czech-written methodology of modern meta-analysis suitable for economics. This part is followed by an applied meta-analysis that investigates the euro effect on common trade exchange, and the analysis is focused on publication bias and the use of the multilevel random effects model. The empirical part is based on 2580 estimates gathered from 33 studies that investigate the relationship between euro and trade volume. The meta-analysis reveals the presence of publication bias, confirms the economic research cycle hypothesis and estimates, according to the available literature, that the true Rose effect lies probably between 2 and 6 percent.
|
guest ::
