Marketocracy Portfolio Update

[zerosum]

A few weeks back, I received a request to post the stock selections I use in my 3 Marketocracy funds. Today I updated my funds, and I like to share with the board the changes I made to the 2020 portfolio, the most aggressive of the 3 funds, as well as some background information.

Here's the description of the 2020 fund from its public Marketocracy page. Link to public page: <http://www.tinyurl.com/9b35z>

**Individual stock selection is based upon statistical analysis of three factor groups: growth, valuation and momentum. Multiple components comprise each group. At the fund's inception, the top 20 ranked stocks from a universe of approximately 2,000 stocks were purchased. Twenty stocks always comprise the non-cash portion of the fund, and each stock constitutes approximately 4.95% of total fund value at the time of its purchase. A stock in the fund is held for as long as it remains in the top 20 ranked stocks from a universe of approximately 2,000 stocks. A stock that falls from the top 20 group is sold and replaced with the highest ranked stock from the current top 20 group that is not already represented in the fund.**

<font class="small">Code:</font><hr /><pre>
Here's the performance history of the selection
process used for the 2020 fund.

1986: 35.4%
1987: 1.4%
1988: 29.6%
1989: 44.8%
1990: 9.3%
1991: 110.8%
1992: 18.9%
1993: 20.7%
1994: 3.5%
1995: 62.4%
1996: 50.9%
1997: 25.5%
1998: 52.7%
1999: 45.2%
2000: 40.8%
2001: 18.0%
2002: -8.1%
2003: 50.7%
2004: 31.4%
2005: 39.2% -- YTD 11/04

Listed below is the current 2020 fund, after today's changes, and the individual buy/hold/sell decisions.

TICKER ACTION

AAPL buy
AET buy
BER buy
CAE buy
ESI buy
GS buy
HAR buy
JWN buy
SNDK buy
STT buy
UFPI buy
WSO buy
XTO buy

BMHC hold
BVF hold
CRDN hold
ESRX hold
FTO hold
LEH hold
VLO hold

BJS sell
BR sell
CKP sell
CMX sell
COG sell
DVN sell
ECA sell
EXP sell
ITG sell
LMS sell
PCP sell
TTI sell
VPI sell

Here's a quick diversification review, based upon Marketocracy's categorizations and current data.

NOTE: The 2020 public page will not reflect the *current* information until Tuesday, 11/08.


Current/previous positions weights by sector/industries (%):

20.6 / 0 Capital Goods
20.3 / 11 Financials
20.1 / 5 Consumer Discretionary
15.4 / 17 Health Care
13.6 / 43 Energy
10.1 / 5 Information Technology

Current/previous position weights by style (%):

25.2 / 24 Large Capitalization
40.3 / 25 Medium
24.5 / 40 Small
10.0 / 10 Micro

54.8 / 44 Value
19.9 / 29 Blend
25.3 / 26 Growth
</pre><hr />

The above information should be enough to generate some discussion. I really don't know if there will be sufficient interest on this board to warrant timely updates going forward. I guess we'll see.

Comments? Questions?

[Uglyowl]

Nice performance! Do you have more specifics on how you find the top 20 stocks?

I don't really know much about the stocks you have, but thought it was great timing to trim the energy sector. Way to ride that out!

I bookmarked your site.

[zerosum]

[ QUOTE ]
Do you have more specifics on how you find the top 20 stocks?

[/ QUOTE ]

I can offer a general explanation.

The selection process is an expected return factor model, where the expected return is relative to the universe of stocks from which the factor measures and sensitivities are derived. Multivariate regression is used to disentangle the interrelationships among factors, allowing for consideration of the *pure* expected return to each particular factor. Each stock's sensitivity to each factor is estimated, normalized and expressed as *B* or Beta. A particular stock's expected return is the sum of its betas multiplied by their respective expected return factor values.

Here's an example, using k factors:

Expected Return = B1*F1 + B2*F2 + B3*F3 + . . . Bk*Fk

The model automatically adjusts to changes in market sentiment regarding the expected return that should be accorded each factor. The underlying assumption of the model is that the present importance of a particular factor is the best estimate of its future importance.

[Sniper]

[ QUOTE ]
The selection process is an expected return factor model, where the expected return is relative to the universe of stocks from which the factor measures and sensitivities are derived. Multivariate regression is used to disentangle the interrelationships among factors, allowing for consideration of the *pure* expected return to each particular factor. Each stock's sensitivity to each factor is estimated, normalized and expressed as *B* or Beta. A particular stock's expected return is the sum of its betas multiplied by their respective expected return factor values.

[/ QUOTE ]

You would have been better served by simply saying you weren't willing to answer the question!!

FWIW, how big of a secret could it possibly be to mention some of the factors you use for your analysis?

BTW, I will take a look at the stocks and give you some thoughts, after I review them. (Thanks again for posting them)

[zerosum]

[ QUOTE ]
[ QUOTE ]
The selection process is an expected return factor model, where the expected return is relative to the universe of stocks from which the factor measures and sensitivities are derived. Multivariate regression is used to disentangle the interrelationships among factors, allowing for consideration of the *pure* expected return to each particular factor. Each stock's sensitivity to each factor is estimated, normalized and expressed as *B* or Beta. A particular stock's expected return is the sum of its betas multiplied by their respective expected return factor values.

[/ QUOTE ]
<font color="blue"> You would have been better served by simply saying you weren't willing to answer the question!!</font>

[/ QUOTE ]

I don't understand why. I can only guess that you may feel that the answer is not informative, or that the answer is too informative.

If you believe the latter, I can assure you that the answer does not provide sufficient information to allow someone to reverse-engineer the process.

I have had business relationships with a number of quant shops that conduct stock selection in a similar manner (i.e., multivariate regression analysis using similar factors), yet produce quite different selection results. On the surface, their processes look extremely similar, beneath the surface, they're decidedly dissimilar. So, again, I'm not worried that someone will be able to take my general description and reverse-engineer my selection process. Its simplicity is truly deceptive.

By the way, I saw your post that incorporated Louis Navellier's recent speculation about the future of index investing. Navellier's qauant shop is one of the shops that I include in the above reference.

[ QUOTE ]
<font color="blue"> FWIW, how big of a secret could it possibly be to mention some of the factors you use for your analysis? </font>

[/ QUOTE ]

Each category contains multiple factors. To mention some, but exclude others, could be misleading. You can identify any number of factors that would reasonably fit into the broad categories of growth, valuation and momentum.

[Sniper]

Hi Zero,

I personally understood what you meant, and why you were being vague/general in your answer... however, it might be useful for the masses reading these forums, if you provided a simple example of the type of analysis and factors that could be used, to help others along in their thinking process.

Just a suggestion.

[zerosum]

[ QUOTE ]
BTW, I will take a look at the stocks and give you some thoughts, after I review them. (Thanks again for posting them)

[/ QUOTE ]

Sniper --

You seem to be familar witn the work of Louis Navellier, so late last week, I ran the stocks I posted in the op through Navelier's stock grader. At that time, Navellier's data was current through November 4, so it coincided with the the time frame of my posted selections. Below are the results, with company name, industry identification and market capitalization columns removed to conserve space.

. . . Sorry, I could not get the format to work. It's really too bad that this forum does not allow the use of some html tags, especially the &lt;pre&gt; tag. &lt;pre&gt; works so much better than the code tag.

Anyway, the portfolio's overall grade was A.

[Sniper]

Yes, the stocks you selected are highly rated on Nav's stock grader, riskmetric and ibd stock checkup [img]/images/graemlins/smile.gif[/img]

Sorry, I meant to post after I did those lookups.