Stock Model

Quote from tradingjournals:

I meant: Are there models (it does not matter on which machine/system they are implemented on) that are commercially viable? I am guessing that the intent of your developing such models is to make money out of them somehow.
More...
Sure. Read the book The Quants. Almost all of the big models are somehow based on these concepts. Commercially available? Hmmm, not sure....
 
Quote from tradingjournals:

How about NLY? This stock is not growth, and I wonder what your models say about it. My interest in this stock is mainly because it a bond-like, has options, and therefore better than a bond (it does not break).
More...

NLY, 4

!!!!Hmm......
 
Model is up 3% since inception. That is an average of 1% a week. There are 40 trading weeks a year. That would be 40% returns before fees.

Standard Deviation will also be interesting. Right now, it is high, but I expect that to smooth out. I don't know what the right number of symbols to hold is. Potentially, I could have a couple of hundred easily, so scaling the strategy to $250M+ is not a problem with no leverage.

Finally, there is still tons of work to do as outlined on a previous post. I have the macro picture, the rankings, but I lack the weights (waaay harder than running a simple covariance matrix), the capacity to simulate different market scenarios, and market timing is still not exactly quantitative (some help from FV, but there are other eigenvalues to consider), etc.

I know I am jumping the gun since "since inception" is only three weeks, LOL, but one begins to see how this is done at RenTec.
 
Just to summarize, here is portfolio, some long some short:

QCOM RVBD AMZN ACOM HTWR IPGP PAY WFC USB ASYS SPRD JCOM AMED MERC TQNT CEPH VPHM DRC APKT CXO LULU ATW CVX IR DECK CRUS LLY AFAM JAZZ TER EIX
 
You must log in or register to reply here.
Back
Top