An Empirical Investigation of the Stability of the Risk Measures of Latin-American Common Stocks through Their Underlying Return-Generating Processes

Authors

Arun Prakash
Raúl Moncarz
Gary Anderson

Date of this Version

3-22-2001

Document Type

Article

Rights

default

Abstract

A recent empirical study (Prakash, Moncarz, and deBoyrie. 2002) found that neither the rates of return probability distribution of the market index nor the individual stocks traded in the Latin American capital markets conform to the Gaussian (standard univariate normal) distributions. Hence, the pricing of assets in these markets may be better modeled through the Kraus and Litzenberger (1976) three-moment capital asset pricing model (CA PM) than through Sharpe's two-moment CA PM. If that is the case, the underlying return generating process will be the quadratic characteristic line model for the three-moment CAPM rather than the linear characteristic line model of the two-moment CAPM. This paper examines the stability of the parameters of these two alternative characteristic line models (which proxy for measures of market risk) over three non-overlapping regimes during the period examined. The study finds that, in general. these parameters are stable for individual stocks over our chosen sample period.

Comments

Prakash, A. J., Moncarz, R., & Anderson, G. A. (2001). An empirical investigation of the stability of the risk measures of Latin-American common stocks through their underlying return-generating processes. Quarterly Journal of Business and Economics, 40(2), 49+. https://link.gale.com/apps/doc/A95447293/AONE?u=flstuniv_sfx&sid=googleScholar&xid=4e443004

Recommended Citation

Prakash, Arun; Moncarz, Raúl; and Anderson, Gary, "An Empirical Investigation of the Stability of the Risk Measures of Latin-American Common Stocks through Their Underlying Return-Generating Processes" (2001). Works of Dr. Raul Moncarz. 2.
https://digitalcommons.fiu.edu/moncarz/2