Direct local linear estimation for Sharpe ratio function

Hongmei Lin, Tiejun Tong, Yuedong Wang, Wenchao Xu*, Riquan Zhang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

Nonparametric regression has been widely used to deal with nonlinearity and heteroscedasticity in financial time series. As the ratio of the mean and standard deviation functions, the Sharpe ratio function is one of the most commonly used risk/return measures in financial econometrics. Most existing methods take an indirect procedure, which first estimates the mean and variance functions and then applies these two functions to estimate the Sharpe ratio function. In practice, however, such an indirect procedure can often be less efficient. In this article, we propose a direct method to estimate the Sharpe ratio function by local linear regression. We further establish the asymptotic normality of the proposed estimator, apply Monte Carlo simulations to evaluate its finite sample performance, and compare it with the indirect method. The usefulness of our new method is also illustrated through a real data analysis.

Original languageEnglish
Pages (from-to)36-58
Number of pages23
JournalCanadian Journal of Statistics
Volume50
Issue number1
Early online date1 Sep 2021
DOIs
Publication statusPublished - Mar 2022

Scopus Subject Areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

User-Defined Keywords

  • Heteroscedasticity
  • local likelihood estimation
  • local linear regression
  • nonparametric regression
  • Sharpe ratio function

Fingerprint

Dive into the research topics of 'Direct local linear estimation for Sharpe ratio function'. Together they form a unique fingerprint.

Cite this