Essays on macroeconomic time series and finance

Resumen

My thesis consists of three parts. In the first part I propose an alternative way to resolve the hours worked -- productivity puzzle that was strongly discussed in the literature during last decade. The debate was generated by the work of Gali (1999), who argues that the empirical response of the hours worked to a positive technology shock is negative that is inconsistent with implications of the RBC models. This result calls into a question the role attributed to a technology shock in the business cycle analysis and ability of the RBC models to explain business cycle fluctuations in the economy. In the second paper I assess the usefulness of the inclusion of the seasonal fractional integration into the empirical model if one wants to estimate the coefficient of fractional integration at zero frequency of a seasonally adjusted series. The third part of the thesis is on the microstructure approach to exchange rates determination. In standard macroeconomic models exchange rate is determined by fundamental factors that are observed by all agents in the economy and constitute public knowledge. In these models, there is no private information and price determination is straightforward and immediate. Unfortunately, their empirical performance is very poor. The underlying idea of the microstructure approach is that the information on the market is asymmetric, i.e., some agents have private information. The bulk of empirical literature has focused on evaluating this hypothesis in a linear framework. This paper analyzes nonlinearities in the relation between HUF/EUR exchange rate and customer order flows.