Estimación Borrosa del Riesgo Beta. Análisis Comparativo

Identificador:  TDX:2791

Handle: https://hdl.handle.net/20.500.11797/TDX2791

Autors:  Laumann, Yanina

Resum:
This thesis represents a contribution to empirical literature on systematic risk at the sectoral level in Latin American emerging markets, by calculating fuzzy betas, sectoral and individual, in Chile, Brazil and Mexico, and to compare its behavior with that of betas in some developed countries as the United States, the United Kingdom and Japan. We propose a fuzzy representation of the model of market that incorporates the calculation of the return of an asset expressed through a confidence interval. With it we incorporate in the calculation of the beta all the information available of the quotation of an asset during the day. As a result of the estimation with that model we obtain a fuzzy beta coefficient. We begin the study by comparing and evaluating the results obtained according to assets return and to the different beta methods of estimation, MCO and lineal fuzzy regression of Tanaka e Ishibuchi (1992) improved with the detection of outliers model of the Hung and Yung (2006). Finally, we advance in the study of fuzzy beta as an indicator of systematic risk. We propose a classification of assets based on fuzzy beta and we verify if two of the traditional hypotheses of the portfolio theory are met in an uncertainty environment: i) sectoral beta shows greater stability than individual beta; ii) the longer the estimation period is, the greater the stability of the beta