Stock Model

Quote from Stoxtrader:

Can you run these cloud (SaaS) companies....

Ariba, Inc. (ARBA)
Citrix Systems, Inc. (CTXS)
Cornerstone OnDemand, Inc. (CSOD)
DemandTec, Inc. (DMAN)
Kenexa Corporation (KNXA)
NetSuite, Inc. (N)
Salesforce.com (CRM)
SPS Commerce, Inc. (SPSC)
Taleo Corp. (TLEO)
Ultimate Software Group, Inc. (ULTI)

...and what is the time period for the model?

Thanks :cool:
More...

ARBA, -4
CTXS, -4
CSOD, -3
DMAN, 0
KNXA, 2
N, 1
CRM , -5
SPSC, -1
TLEO , 1
ULTI, -3

There is no specific time period. The ranking can change from day to day, but the rankings tend to be very stable. I have only been running the model since the individual stock correlation with the overall market is very high. In that scenario, it makes sense to rework the rankings say every 30 handles that SPX moves in one direction or another. Similarly, if the stock moves strongly in either direction from the date the ranking was computed, it makes sense to run it again. The point is that the ranking should be run whenever it makes sense to do so, but at least once a week.

When (if ?) stocks go back to having a more self-reflexive existence where they are not so dominated by the overall market, I suspect this model will become even more valuable in that environment.

It is important to note that this model is only half of my strategy for putting on positions. The other half is how to build a book of stocks (long/short for me but the variation of possible strategies around a sound ranking are huge) around a set of the rankings. As I have mentioned, because the correlation with SPX is very high right now for stocks in general, I use my other major thread about SPX levels to aid me in "leaning" my book (long or short), although I am far from perfect in doing that. Leaning a book by itself is a large theoretical mine field, as it can be done so many ways (net beta, net dollar, net long/short, risk-reversal bias, dispersion, etc etc etc...)

Finally, my overall strategy will evolve to where the book will be on [an] efficient frontier. This is still not implemented, I sort of do it by feel and I am pretty decent at it (after many years of doing this, you get a feel where (what sector etc) emotion runs highest), but I would like to be 100% quantitative.
 
Quote from nitro:

the individual stock correlation with the overall market is very high

the correlation with SPX is very high right now for stocks in general

[/B]
More...


Thanks for the details. I would expect stocks to be highly correlated with the index they're in, for example Apple and Google should be highly correlated with both the Nasdaq 100 and the S&P 500. However Apple and Google should be less correlated with the Russell 2000.
 
Quote from Stoxtrader:

Thanks for the details. I would expect stocks to be highly correlated with the index they're in, for example Apple and Google should be highly correlated with both the Nasdaq 100 and the S&P 500. However Apple and Google should be less correlated with the Russell 2000.
More...
Right, I mean that individual news/fundamentals in the stock is often trumped by the direction of the overall market (which I am using SPX as a stand in) in the recent past.
 
Stock Model new equity high. Not surprising since the book is leaning short. It is running at about 7% every quarter. Assuming it can be sustained, I see no reason why 28% a year without leverage or something close is very possible.
 
New equity high again, this time the gains being led by the short positions in DECK and PAY. Nothing fascinates me more than when my position (book) makes money when the market goes up or down.
 
You must log in or register to reply here.
Back
Top