Generally, I don't avoid them in my own trading. Some systematic traders build their own exclude list over time. Mostly manually.
All months positive so far YTD. This year I turned off one strategy. Rationale: it was long and hit max DD while market was hitting new highs, which haven't happened before. Ended up being Net ~0$ on it, as it made good money at the beginning of the year. Will follow up on it monthly. Have couple of others I might bring back online this summer. At the end of April used discretion to reduce position sizes for longs by 30% and increased all shorts by same amount. Converted ~70% of capital to CAD. So far all discretionary decisions paid off. Overall I'm happy with all strategies. Retirement account ones (on the right) have less turnover & no shorts = worse Risk Reward, as expected. Green/blue - long strategies, otherwise - short. Gray (barely visible one) is combined equity from all per account. Out of ideas on what to write about hence been quiet here. Busy with technology business otherwise. We are are ~120 people now, about doubled over last 12 months. Moved to a place just by a beach and been spending way more time outside (as not it is sunny and nice again in Vancouver). View from my desk. Val
Congrats on your success so far. What most do not realize is that successful algotrading is not too exciting. No news is good news. If you don't mind me asking, what kind of technology business are you working on now? 120 people sounds like a serious operation.
Can you elaborate on how do you handle corporate events? Say you are short rights due to a rights issue on a stock which you have shorted? There is no liquidity in the rights and you have might have to pay a lot to cross the market.
Great to find this post! I've spend hours reading it today. The book Unknown Market Wizards that I'm reading leads me here eventually. Learned a lot and am sure will try to explore my own MR strategy to add to my current couple of trend following strategies that I've been trading for the last couple years. Thanks @ValeryN
It has something to do with disrupting how business banking is done. We offer it as SaaS to banks (our customers) where end users are businesses.
The only thing I do that considers corporate events is having my exit management split/ticker change neutral. Theoretically, worst case would be being short of something that is halted for months due to fraud or gets delisted (likely=getting stuck paying insane borrow rates), but I don't short penny stocks which significantly lowers risk of getting caught in that. About once a week I look at a list of open positions and their age. If an unknown unknown would happen, theoretically, I will spot it there and might need to contact my broker.
I'm happy you found it useful. Thank you. Market Wizards is a great series for inspiration and learning about different styles. I really like earlier books too. For initial MR strategies ideas I would recommend to look at Howard Bandy's book on Mean Reversion and Laurens Bensdorp's Automated Stock Trading systems. I've posted an example of at least one system here too, as well as some examples on how those systems get destroyed by hard stops and associated costs. This journal has too many pages now, even I struggle to find stuff. Maybe will move it to a blog or something so everything is a bit more structured. Open to ideas.
Hey! Quick question. Do you feel like there is any added-value to test with ddjusted (for splits and dividends) EOD pricing data? Thanks in advance and solid trading!
Yes, either that or having your backtest engine account for that, assuming dividend & split information is coming from a data source. Otherwise you will see unrealistic gaps, they will look your test results better or worse depending on a strategy. When in reality there was no gap.