Document Type

Working Paper

Date

2001

Embargo Period

6-28-2012

Keywords

time-series models, dynamic quantile regressions, foreign exchange, international economics, agglomeration heteroskedasticity models

Language

English

Disciplines

International Economics

Description/Abstract

This paper proposes a robust estimation procedure, the bounded influence estimate (BIS), which is robust against departure from the conditional normality of the autoregressive conditional heteroskedasticity (ARCH) models to describe the behavior of exchange rates. First, the BIE identifies the additive outliers (AO, e.g., Fox 1972) caused by abnormal information arrivals which may be triggered by changes in domestic policies and international shocks. Identification of outliers allows us to analyze the major economic and political factors that contribute directly to the dramatic changes in exchange rates. Second, the performance of the BIE is compared with the maximum likelihood estimate (MLE) and a semi parametric estimator (SP) of Engle and Gonzalez-Rivera (1991).

Additional Information

Harvest from RePEc at http://repec.org

Source

Metadata from RePEc

Share

COinS