Forecasting Intraday returns using Kernel regression / Previsão de retornos intradiários através de regressões usando funções-núcleo

AUTOR(ES)
DATA DE PUBLICAÇÃO

2007

RESUMO

The contributions of this paper are twofold. First we discuss and apply a method for the eva luation of non linear regressions in forecasting intraday returns of Brazilian stocks, with the aim of maximizing the return of a simulated trading portfolio. Second, Kernel regressions associated with Nearest Neighbors sample partitioning are carried out. Some independent variables are technical indicators, which parameters are optimized in-the-sample. The results are positive as a trading strategy and outperformed the linear autoregression benchmark model in a quantile per quantile analysis

ASSUNTO(S)

vizinhos mais próximos kernel regression regressão usando funções-núcleo retornos intradiários technical indicators economia indicadores técnicos intraday returns nearest neighbors

Documentos Relacionados