[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[RT] HeatMap/HiMarq model portfolios



PureBytes Links

Trading Reference Links

>The HeatMap has gone further and is trading a strong-sector-strong-stock
portfolio starting Jan 3 and results will be available for >public scrutiny.
I will get you the url once I have it.

Here it is:

The article:

http://moneycentral.msn.com/articles/invest/models/4979.asp

The Himarq portfolio:

http://moneycentral.msn.com/articles/invest/models/himarq2000.asp

The annual calendar (same as Heat Map without all the clutter):

http://moneycentral.msn.com/articles/invest/models/HIMARQ_CAL.ASP

As you will read in the article there will be certain changes made as we
gain more experience re seasonality.

Currently the major area under review is the weight to be attached to an
index/stock that tends towards a 100% strike rate in terms of frequency of
recurrence with a low standard deviation and >5% returns for the month.

The HeatMap in its present form looks at absolute returns - thus
- if an index has a 10% avg gain for a month over a 10 year history and
- 5 years out of 10 positive, with the other 5 negative

It gains precedence over
- an index that has 5% avg for a month over the same 10 year period
- but all 10 of the years closed positive.

Intuitively, one should attach more weight to the latter case.

We have tried to address this in the equities you see on the Himarq model
portfolio page, the indices are a work-in-progress.

The indices' ranking for Jan is in the gif attached with this email, which
is the HeatMap for January using Dec 1989 as base year - ranked by % months
positive. As you will see, the Top 3 indices (blue lines = HeatMap blue
cells) are not the same when weighted with historical recurrence.

Comments and suggestions are welcome. All content on Microsoft Investor is
copyright Microsoft and its authors. I am independently trading this
portfolio using a blend of technical/option techniques and will, from time
to time, have positions that deviate from the model either in choice of
security, or period held, or amount invested. The core logic will remain
untouched.

This is not a recommendation to buy or sell any securities. Readers are
strongly advised to do their own research before committing capital to a
trade.

Gitanshu

PS: I owe replies to a lot of folks who have emailed me privately. I wanted
to meet this article's deadline, and will now try to catch up with you.
Thank you for your patience.

Attachment Converted: "f:\eudora\attach\MthPos.gif"