Social and Behavioral Sciences Commons^™

Fractal Structure In The Chinese Yuan-Us Dollar Rate, Raul Matsushita, Iram Gleria, Annibal Figueiredo, Sergio Da Silva

Sergio Da Silva

Price changes of the Chinese yuan-US dollar rate are found to display a Sierpinski triangle in an Iterative Function System clumpiness test. This fractal structure commonly emerges in “the chaos game”, where randomness coexists with deterministic rules. We show that a threshold model with four states, two deterministic and two stochastic is able to replicate the properties of the yuan-dollar changes in general, and the Sierpinski triangle in particular.

Classroom Guide To The Equilibrium Exchange Rate Model, Sergio Da Silva

Sergio Da Silva

The article presents a classroom-suited version of the equilibrium exchange rate model of Stockman (1987) that features Cobb-Douglas functional forms for both production and utility, and considers foreign exchange intervention explicitly.

A Brief Overview Of The Current State Of Exchange Rate Modeling, Sergio Da Silva

Sergio Da Silva

The paper provides a brief overview of the current state of exchange rate modeling.

Chaotic Exchange Rate Dynamics Redux, Sergio Da Silva

Sergio Da Silva

This article generalizes the results shown in De Grauwe, Dewachter, and Embrechts (1993) in a more sophisticated framework. In their model, the speculative dynamics resulting from the interaction between chartists and fundamentalists are incorporated into a Dornbusch-style model to generate a chaotic nominal exchange rate. Here the model of Obstfeld and Rogoff (1995, 1996) replaces the Dornbusch model, and chaotic solutions are still shown to be possible for sensible parameter values.

Chaos And Exchange Rates, Sergio Da Silva

Sergio Da Silva

This paper surveys the literature on chaos in exchange rates.

The Role Of Foreign Exchange Intervention In A Chaotic Dornbusch Model, Sergio Da Silva

Sergio Da Silva

Adaptive Estimation In Timeseries Regression Models, Douglas Steigerwald

Douglas G. Steigerwald

I develop adaptive estimators for linear regression with serially correlated errors. The efficiency results hold even when the serial correlation structure is unknown. Simulations indicate that efficiency gains can be substantial with samples of only 50 observations. We apply the method to a study of forward exchange rates.