This paper investigates the time series properties of the return of American Depository Receipts (ADRs) issued by Latin American companies. Specifically, we examine the presence of nonlinear trends and chaos. The motivation of the paper is twofold: First, we use a micro data set, wich can avoid the problems encountered by previous research of nonlinearity an chaos in micro financial data. Second, we choose returns of Latin American ADRs because of its unique characteristics, which might make them to behave differently from the U.S. capital market as a whole. Even though the powerful BDS statistic strongly suggests the presence of nonlinearity in the Latin American ADRs returns. The BDS statistics, applied to the standardized residuals of the EGARCH model, rejects heteroskedasticity as the cause of nonlinearity. On the other hands, the locally weighted regression indicates the presence of chaos in the Latin American ADRs returns. Our results are relevant because previous research has failed to report evidence of chaos in stock returns.
Keywords:
American Depository Receipts (ADRs), Temporary behavior, Latin American companies
Urrutia, J. L., & Vu, J. (2001). Analysis of the temporal behavior of Latin American ADRs. Estudios De Administración, 8(1), 1–17. https://doi.org/10.5354/0719-0816.2001.56756
