Here are the rules.
Using the Nasdaq 100, select the 5 stocks with the highest rate-of-change over the last 252 bars (roughly one calendar year) and buy them at the close 5 days before the month’s end. Sell on the close of the 1st day of the next month.
Seems simple enough, no?
I’m testing from 1/1/2000 to 10/6/2010. I’ve not included any commissions or slippage.
The Results:
Equity Curve:
Profit Table:
Summary:
Not bad, eh? Did we just get lucky? It is really too early to tell. There are some limitations/caveats that should be discussed.
1. The Nasdaq 100 database is accurate for the present but would have some different stocks in the past that are not included in the current construction. I could recreate each year’s construction to match the addition/deletion process and then use them to test over, but I didn’t. Perhaps I will do that in the future. It would take a fair amount of time.
2. We are looking at the best performing stocks of the past year and these stocks continue to exhibit momentum. This means that often, for months in a row, the same stocks are being bought. For example, BIDU has consistently been at or near the top of the list for 2010. This means that BIDU was one of the 5 stocks purchased every month-end for most of 2010.
3. We might have just gotten lucky. To truly determine if we are capturing seasonality, or just catching 6 days worth of momo from the best stocks of the past year, the tests will have to run through every day of the month, holding for six days, and then comparing results. Even doing that, we may never be sure that we aren’t just capturing some momentum.
Where to Next?
There are many variations on this theme. For example, what if we bought the worst performing stocks of the Nasdaq 100? (Spoiler Alert: Results improve). What if we shortened the look-back period? Used stops? Filtered the picks for volatility?
Before we get to that fun stuff, I’m going to try and figure out if we are catching seasonality or momentum.
Thanks to Howard Bandy for his help with the code for this project.
Here is the next post in this series: Testing End-of-Month Markup with a Survivor Free NDX Database.
Ha! A little crazy, since you will not know what stocks will be best in the future.
How about doing a more balanced study? Check out qqqq and spy.
Huh? The Nasdaq 100 are the components of the QQQQ. I could easily extend it to the components of the S&P. I don’t understand your comment about not knowing which stocks will be the best in the future??? Balanced study?
Survivorship bias. Poor performing stocks have been dropped from the index over time.
Tom, why then does performance improve when we choose the worst stocks of the past year?
I’m not saying that there is not survivorship bias. I listed it as a caveat. I’m just not sure that there is much difference between the 100th stock and the 101st stock on the Nasdaq, except the 101st is not part of the Nasdaq 100.
The link below shows the additions and deletions over the years, in case you are interested.
http://www.nasdaq.com/indexshares/historical_data.stm
If this really holds up, you’ve got the lowest risk system of which I can think, as you’re not hanging out in Mr. Market for very long periods of time.
Sort of. Risk is not necessarily equal to exposure. It is also related to volatility.
You could be 100% long a bond etf and have less risk than being 10% long a portfolio of stocks.
I am interested in the raw overall market data during the same period.
You mean like what an average stock did during the same 6 days?
Nice.
Another parameter to mess with is filtering over 252 bars. The fact that we saw BIDU pop up is that it continued to stay near 52 wk highs all through the trend. What about a smaller timeframe, or incorporating change in volume?
Also, I’m surprised that a long-only system did so well in 2008, but that’s probably an anomaly due to the lack of time in the market.
We definitely want to look at a smaller time frame. I’ve been playing around with the best 5 stocks of that month, and it works. Plus, a smaller time frame may reduce some of the momentum effect and highlight any seasonality.
In 2008, I think you’ll find that the market had some BIG days at the end of the month. Chicken or the egg? Who knows.
Very cool study. 2 comments:
1) How much of an affect does the nearly 8 years of a phenomenal bull market have?
2) How tech heavy is the nasdaq 100 and how much of an affect does that have?
1) That bull market was, by my rough estimation, from 2003-2007 and again from 3/2009 to 3/2010. Or roughly 6 years out of the 10 tested. I think you’re looking at it wrong. The bear market was much more phenomenal than the bull market.
2) A couple comments above I give a link to the nasdaq site. It should answer your question about how tech heavy the index is.
One way to get at what the various Flies are asking is to make each trade hedged by the QQQQ — i.e., measure excess return. That eliminates market seasonality.
Thanks for translating.
I’ll run that too for comparison.
Nice work Woodshedder!
How is the performance if separated into Bull and Bear periods?
I guess one can approximate that from the monthly returns.
You beat me to it.
What is the return for the QQQQ’s traded the same way?
Aris, hold on. I’ve got access to a survivorship bias free Nasdaq 100 database. Let me run the tests over it and then we’ll talk more.
Do you live in S.C., Wood?
Born and bred, CoC grad, but live in Richmond, Va. now. Why?
My fund has an office in Charleston. Was there in April.. Very nice.
Spent many good years there. Lived downtown on the peninsula in college and then on James Island and Folly Beach. Left to go to grad school, and never came back. 🙁 My wife is more of a mtn. girl, but she is beginning to like the beach. We were at Folly for a week 2 weeks ago.