Due to the following reasons， I only implemented the HAC method in the paper.
1 Bootstrap method performance is very poor
2 The bootstrap method relies on automatic calculation of the optimal block size, which is difficult to implement
3 Due to time, the HAC method based on the prewhitened QS kernel has not been implemented either
As mentioned in this issue： How can I build my confidence that my strategy returns better, I implemented my first version.
This version basically refers to the author's matlab version of the code. I put the original matlab version of the code at https://github.com/weiguang-zz/robust_sharpe_test. the paper reference: http://www.ledoit.net/jef_2008pdf.pdf
Due to the following reasons， I only implemented the HAC method in the paper. 1 Bootstrap method performance is very poor 2 The bootstrap method relies on automatic calculation of the optimal block size, which is difficult to implement 3 Due to time, the HAC method based on the prewhitened QS kernel has not been implemented either
thank you