Does anyone here try to trade discount mean reversion of ETFs? There was a 2018 paper Inefficiencies in the Pricing of Exchange-Traded Funds by Antti Petajisto. High frequency trading firms will arb the ETFs vs. the underlyings, but my question is whether retail traders who trade maybe daily and hold for a few days until the discount mean-reverts can achieve a good Sharpe ratio, especially if they don't hedge.
Better to spread the stocks/ETFs and trade the outperformance. Sector ETFs with differential beta can be traded long/short using volatility and risk ratios. Trading the ETF against the underlying can be done, but a perfect hedge will not be profitable, in general. You will pay continuously for perfect hedges and you are not executionally advantaged in that endeavor. Cross-hedged ETF exposure and differentials of baskets of ETFs can have favorable risk/reward characteristics. Also, you can create synthetic indexes using sector ETFs (using beta and vol).
the ETF and the constituents are never out of line. And if you calculate iNAV realtime and think there is an arb there's no liquidity or cost prohibitive.