I just did a ranking on all the stocks of the S&P 500 and it turns out that only 483 stocks came up. Is there a reason 17 stocks are missing from the S&P 500 universe? Does anybody know which 17 stocks are missing?
Doug
Doug,
Here is a link to the S&P500 list:
There are 500 stocks in Standard & Poors’ list.
Here is a Link to a S&P 500 Ranking System:
http://www.portfolio123.com/rank_details2.jsp?rankid=15572
This Ranking System ranks the stock universe in order of S&P 500 stocks by Market Cap first, and follows that with all other stocks by Market Cap. If you run the Rank test you will see that the first 483 stocks are in the S&P 500.
You’re right Doug. Where are the other 17?
Denny 
Denny,
Thanks for replying to my post. The reason I’m concerned about this “missing 17” is because every test I’ve run on the S&P universe doesn’t seem right. Specifically, all the rankings and simulations seem to be biased towards trouncing the S&P. I know the S&P is weighted and the buckets are not weighted, but the differences in returns for apparently arbitrary factors seem wrong.
For example, I did a test just on raw prices. Amazingly, buying S&P stocks in a variety of arbitrary price ranges, and simply selling 4 weeks later, beat the S&P index handily.
I’m not sure what the missing 17 stocks are but isn’t it possible that their absence might be skewing the results?
Doug
Doug,
Till last week they were all 500 stocks listed for S&P 500. It is only this week that there are 17 missing. They all aphabatical by symbol - CTX
CTXS
CVG
CVH
CVS
CVX
CZN
D
DCN
DD
DDS
DE
DELL
DG
DGX
DHI
DHR
I do not know why they are missing from the S&P 500 universe because they do have ranks available for last week. This just happening in the latest week cannot explain the results you are getting.
KJ
Don’t know if this is the issue on the arbitrary returns - but I do know the top 50 companies in the 500 make up 80% of the index. Because it is a cap-weighted index, the large companies dominated.
I also know that the bottom 450 companies performed much better over our testing period than the top 50 as small/medium caps have outperformed large-caps by a large margin.
For example, while the SP500 Index is just now getting back to its level of 5 years ago, the SP400 Midcap Index is up over 50% in that period and the SP600 Small Index is up over 75% for the period.
I ran a backtest from 3/31/01 in the screener with Universe(500)=1 as the only criteria. The average return (which would be equal weighted as opposed to cap-weighted) is 5.7% versus 1.6% for the S&P. That corresponds with past research I have done and other research I have seen. In general there has been a 3-4% bias in average return of the companies vs the index.
This is an issue I have brought up in past posts - we really need to have the Russell 3000 index in P123, although it also is cap-weighted.
For myself, I generally don’t assume a sim has excess returns unless/until it is well over 20% annualized given the upside down nature of returns in the market the last 5 years. A blind monkey can create a sim that would beat the S&P over the last 5 years because the largest 50 companies have been the worst equity investment possible over the period.
So I don’t think there is a systemic data problem, just looks like the missing 17 are from the weekend data load given they are consecutive alphabetically.
The sp500 universe has been corrected. It seems we got bad weekend data. Thank you for reporting this. It will be available when we reload the servers. I’ll start the reload in a few minutes, and should take about 20 min.