Hacker News new | past | comments | ask | show | jobs | submit login

This section tells me that the data is insufficient to draw conclusions:

> The strategy goes as follows: we purchase SPY on a new moon, and re-sell it on the next full moon. And repeat that every lunar month.

> Clearly, it fails at beating SMA crossovers. Or at doing basically anything, an investor using the moon phase strategy starting with 10.000$ would end up with only 11.110$ and a Sharpe ratio of only 0.09.

Under EMH, technical analysis doesn't work, and under technical theories there's no reason to expect a wax on / wane off strategy to do anything. Our strong assumption should be that a strategy that has you in the market 50% of the time should produce 50% of the return with 50% of the variation.

However, this arbitrary control strategy instead provided a strong return difference, somehow avoiding most periods of growth. If this backtesting period gives a false positive signal that this strategy differs from no strategy (appropriately weighted in-market investment), then the period is also likely insufficient to confirm the strength of the crossover technical strategy.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: