search

lambdaclass / options_backtester

Simple backtesting software for options
MIT License
161 stars 29 forks source link

Compare results with data we have from books #1

Closed unbalancedparentheses closed 4 years ago

unbalancedparentheses commented 5 years ago

For now we have this data to check that the backtester is working fine. I will start uploading other data that I have.

Screen Shot 2019-03-21 at 5 04 32 PM Screen Shot 2019-03-21 at 5 04 40 PM
unbalancedparentheses commented 5 years ago

I would take the sentence "even the inclusion of the 1987 crash failed to make put buying profitable" with a grain of salt. DTE, % OTM, VIX general market valuation(CAPE, Market Cap to GDP, Tobin's Q ratio, etc) should be taken into account to affirm that a strategy is profitable.

Anyway, I agree that in general buying insurance is not profitable if you are measuring the returns of the insurance per se. But doing so can make the ensemble be more profitable, specially when the insurance has a explosive payoff.

Screen Shot 2019-05-11 at 23 10 14

Taleb's ideas expressed in many interviews (http://www.forex-ratings.com/forex-interviews/?id=12083 and https://merage.uci.edu/~jorion/oc/ntaleb.htm) , Universa's brochures/papers and "Insurance makes wealth grow faster" paper are a great take on this.

I am mentioning this since most option backtesting strategies are done in short periods of time and only check the % of time that the strategy wins. In short timeframes (less than 10 years) selling volatility in general is profitable until it isn't.