Bridging emerging market models and investors realities: the case of currency and external debt markets.
MetadataShow full item record
Our target is to objectively quantify important aspects of emerging economies’ financial markets and deliver value adding actionable recommendations that can be used by a wide spectrum of end-users like academics, policy makers and real life investors. We create two quantitative models that capture the dynamics of global Emerging Market currencies and sovereign debt ratings. We build on the extensive literature on Emerging Market crises and introduce a number of methodological and conceptual innovations. A wide range of market stylized facts and practical and intuitive limitations dictate the way we progress with our research, from considering and selecting dependent and explanatory variables to the way we apply and interpret the model results. We first estimate a parsimonious panel specification that models and forecasts Emerging Market currency dynamics and produces trade signals for investing in one-month forward exchange rates. The second instrument models and forecasts credit ratings assigned by two of the leading rating agencies to Emerging Market sovereigns. The specifications we select are tested on the basis of their statistical and forecasting performance which is found to be solid and unbiased. The currency model is further tested based on its ability to generate profit making trading portfolios. The ratings model is also assessed based on its forecasts for forthcoming sovereign rating actions. We proceed to apply both models on real time data and compare the results from blindly following the model recommendations to a situation where an investor filters these results by superimposing his market awareness and subjective judgement. Our findings suggest that the tools developed here can reliably be integrated in an investor’s decision process. The events of late 2010 suggest that many of the ideas presented in our work can be implemented to Developed Markets and be expected to produce interesting and usable results.
- Theses