Pregled bibliografske jedinice broj: 156940
Let Figures Speak:Meta Regression Analysis of Trade Effect of Exchange Rate Variability
Let Figures Speak:Meta Regression Analysis of Trade Effect of Exchange Rate Variability, 2003., magistarski rad, Stefforsshire University, Economic Division, Stoke on Trent, Engleska
CROSBI ID: 156940 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Let Figures Speak:Meta Regression Analysis of Trade Effect of Exchange Rate Variability
Autori
Bruno Ćorić
Vrsta, podvrsta i kategorija rada
Ocjenski radovi, magistarski rad
Fakultet
Stefforsshire University, Economic Division
Mjesto
Stoke on Trent, Engleska
Datum
01.10
Godina
2003
Stranica
67
Mentor
Geoof Pugh
Ključne riječi
Meta analysis; Trade effect; exchange rate variability
Sažetak
The trade effects of exchange rate variability have been an issue in international economics and policy making for the past quarter century. Yet, so far, neither theoretical nor empirical work has converged towards consensus. The contribution of the paper is to apply meta-regression analysis (MRA) to the empirical literature. This demonstrates that on average exchange rate variability exerts a statistically significant negative effect on international trade. In addition, MRA explains the wide variation of results in this literature – ranging from significantly positive to significantly negative effects – and suggests new lines of enquiry. We find that different measures of exchange rate variability may influence the estimated trade effects, but not in any systematic manner. Negative effects are more likely when trade is beyond the range of forward markets and other methods of managing exchange rate uncertainty and are most likely to be identified when researchers focus on low-frequency exchange rate variability. Our results also suggest a regime effect, whereby the trade effect of exchange rate variability is conditioned by the institutional environment. So far, this effect has not been recognised either in the theoretical or in the empirical literature. Finally, the paper makes a recommendation for the conduct of MRA: the grouping of results by study suggests that weighted and cluster-robust estimators should be used to complement OLS estimation, which hitherto has been the norm in MRA.
Izvorni jezik
Engleski
Znanstvena područja
Ekonomija