Hello, I am looking for a high quality options backtesting software that is able to backtest not only P/L, but also Greeks and draw historical options payoff charts. I am currently using PaperMoney from ThinkOrSwim which is ok, but one can only backtest P/L in there and not Greeks. I know that LiveMoney is able to do that, but I am not an American and TD Ameritrade doesn't open live accounts for Europeans. Any advice or shared experience is appreciated. Thanks. Luci
Hi Luci, I know your post is a few month old but i was looking for exactly the same and found http://www.oscreener.com It allows to backtest option strategies using greeks and stock technical performance parameters! The only problem they don't have all the option strategies I wanted to backtest. http://www.oscreener.com only supports backtesting of Bull Put Spread, Bear Call Spreads and Long/Short Put & Call options. Anyone knows a good free or inexpensive tool to backtest straddles and calendar spreads?
I would suggest purchasing your own historical options data and just use either R or python/pandas for backtesting. Almost every single software I've seen incorrectly calculates greeks/iv anyway.
Do it yourself. Not found anything decent that also allows trading then. JUST TODAY I put the first 38 cores of my cluster online, distributed over various machines. It is nice when you can do a backtest in minutes, and a full optimization in more than an hour, with full tick replay, bid, ask, all that crap. And KNOW the results are good.
I am looking for a options backtesting platform that allows me to backtest options spreads and complex entry / exit / adjustment criteria. I would like to backtest strategies such as covered calls, strangles, butterflies, and iron condors. And I would like to be able to create trading rules and adjustments based on criteria such as option greeks, volatility, technical studies, return on investment, etc.. I am currently using OptionVue & ToS. But these tools are ill-suited for these types of analysis. The process is very painful and manual. Is there a better, more automated way to perform these types of options strategy backtesting?
I had the exact same problem. I tried OV and ToS, but couldn’t get the results I wanted. So I downloaded historical option price data and used Excel & R to do my analysis. That proved to be much more difficult than I anticipated. Recently, I started using OptionStack to backtest my options trading strategies. It works great! Within a few minutes, I was able to create and backtest my trading strategies across years of market conditions. In fact, an iron condor strategy that had taken me several days to manually backtest took less than a minute to backtest using OptionStack. But the only issue is that the platform is currently in beta, and so the data set is limited.
I've tried optionstack briefly but its hard to build custom strategies using it. For example try selling a straddle or an IC once every 15 days and adjust when say a wing is hit. Not sure but doesnt seem you can test custom strategies with it; only canned technical studies based ones. Maybe its possible if you write scala code but its not obvious that its possible. -gariki
Well, I've written a backtesting software of my own and I'm looking to expand it now into an automated trading platform. I've purchased one year of options data from https://www.historicaloptiondata.com/ and so far tested about 15 underliers, some 10 equities and 5 indexes / ETFs. Results are mixed: while it looks like I'm definitely making money, it seems no model / strategy works on all underliers. Like something that works great on SPX, it's just plain random if not straight losing on WLL, and so on. For trading, I'm using Interactive Brokers. So far I'm in read-only mode, getting marketdata automatically in my program but don't dare send orders automatically yet. So I'm looking too, for someone or a few people to work with me on the software (Java) and algorithms (quite a lot of quant stuff / math) and improve it. I also got a business plan, but in the first phase I'll trade by myself and so have to do my possible partners. Anyone interested?
Got a new idea and have to ask you what you think of it In short, I'm thinking of open-sourcing my backtesting software. I ran a first series of tests with reasonable results, entered a first real trade based on them and so far it's OK. And I realized I gotta change the algo a bit, I feel this will get better results. Based on previous experience, this might be just the point where the rocket really takes off or it might crash again. And if it crashes again, I dunno if I got the strength to carry on - all by myself. It would be really useful to have others to talk about models, architecture, help with implementation. But the only way I feel I could get enough people interested would be to open-source the stuff, which to a large degree would render it useless. Or so I thought, until I got this new idea: what about instead open-sourcing the code, I'm just publishing and discussing the specs? The WHYs (markets, models) and WHATs (architecture) instead of HOW (implementation). Then, people can just go on and do their private implementation or contribute to the open source project. We could discuss / refine the specs to a re-designed version, which may be similar to what I already got or it might be very different. For instance, one software limitation I've got currently is that the stuff only runs on a single PC, so even on a core i7 with 8 threads it takes a lot of time for the backtests to complete. I definitely have to parallelize / distribute that and here there could be interesting discussions about weather to use something like Apache Spark or implement something custom (I would have to distribute proprietary code and license-restriced market data, and would prefer to use SQL). Discussing the models would take some math knowledge: differential equations and martingales but if people have the background, that can be taught. I'm thinking of a series of tutorials which include math + programming and actually explain / make people figure out what this option pricing is about. Black-Scholes model, delta-hedging, delta-gamma hedging, an(other) explanation for the volatility smile... So what do you think?