I would like to see someone test random entries into the universe of equities, with maybe only a volume filter, and then apply differing money mgmt condition on the entries, testing the effects. I have a feeling they would not perform nearly as poorly as most would imagine.Not one to turn down a challenge, and sharing a curiosity on the subject of randomness and stock trading, I have decided to take up the task.
My first order of business was generating a random list of stock symbols to trade. A quick search for random number generators took me to Random.org, a website that generates true randomness using atmospheric noise. Huh!? Just read this.
After poking around the website for a couple minutes, I noticed the List Randomizer, which will turn 10,000 items into a random list, all I had to do was provide the raw material in the form of stock symbols.
To keep the experiment as random as possible, while also being realistic enough to trade, I had to select stocks with adequate liquidity. So, I limited the stock universe to issues with a price between $10 and $500, with daily volume between 500,000 and 100 million shares. These two criteria produced a target list of 1,249 stocks via the stock screener at MarketWatch.
I entered the symbols into the list randomizer, which turned my alphabetical list of symbols into a random list in less than a second.
The next step took me to Investopedia's stock simulator. The simulator provides basic tools for paper trading stocks and options, and includes a $100,000 starting balance for portfolios.
To keep this experiment simple, I will be buying $10,000 worth of stock in the first 10 stocks on the random list. Each stock will be bought using market orders at the open. Each stock will be held for five days, and then sold at the open on day six. A stop loss of eight percent will limit downside risk. Stocks that stop out will not be replaced until all existing positions are closed after the five-day holding period.
Now that the ground rules have been established, let's meet the first 10 random candidates. Market orders have been placed for the following stocks: GNK, BMS, BCS, DBRN, ETR, DRIV, STZ, BDX, HAS, and NDAQ.
7 comments:
Very cool Dog. I'm going to try and think about how to track different money mgmt. techniques, on these randoms.
You want to share the list of symbols? I can generate the list each week, email it to you, then we can each post on the topic using our separate position sizing/money management method.
Let me know if you want more of today's list. I can send the entire file to you via email if you wish.
i think you need to add 10 for the short side as well so mkt direction wouldn't affect
Gilad,
Short entries would give a more complete picture, but since most retail traders/investors only play the long side, I think it is okay for the purposes of this test.
Also, most of my swing trading strategies are long only, so doing a long-only random entry experiment will allow me to compare the results of to the performance of my strategies, as well as the overall market.
I'll keep the short idea in mind, however, if I do another experiment.
Dog, that sounds like a good plan. Send it on over, rover.
Yes the stop loss acts as a Maxwell's Demon (Wikipedia it). Ive had the same thought. Will be watching.
post for follow-up comments
Post a Comment