Abstract (Translated):
In this article, the behavior of the returns of some assets of MILA is analyzed, with the objective of looking for evidence of persistence and evaluating the impact of their presence in the decision making of investment portfolios. The methodology of the rescaled range is used in the estimation of the Hurst coefficient as a measure of persistence and the results are verified with the adjustment of Anis and Lloyd and the estimation of Higuchi. An inferential process is added to the Hurst coefficient for each of the assets analyzed. The performance of portfolio optimization including estimates of persistence and the results of its inference were compared with independently optimized portfolios. A better risk-return relationship is observed by including the pattern of persistence, only when the inference is supported by evidence.