“ We need to show that machine learning beats linear regression on financial data”

Are there any hedge funds that use linear regression? Most models are non-linear based on factors or even TA indicators as far as I know.

At any rate, there is a certain level of data-snooping bias in your backtests, i.e., Improving the model by reusing the same data. But at the end actual performance will be the judge and I hope it will be good. The main problem quant funds face is not the models but lack of inflow of dumb money to profit from.