On Mon, 26 Feb 2007 18:27:43 -0600, Tad Borek <borekfm[at]pacbell.netwrote:

- quote -

The MIFP charter speaks to this subject. I quote from charter item
#8:

"8. Well-tempered disagreements and constructive criticism to
participants of this newsgroup are acceptable in the context of
adding new information to the discussion. Harassment, personal
insults, attacks and profanity towards group participants will
not be tolerated whatsoever."

According to the online dictionary, "harass" implies systematic
persecution by besieging with repeated annoyances, threats or demands.

IMO a key word here is "repeated".

If anyone feels that they have experienced such treatment on this
newsgroup, please bring it to the attention of the Moderators at their
private email addresses. For more, see the weekly post, "Posting to
misc.invest.financial-plan".

--------------
As usual, please do not respond to this post. Comments on newsgroup
policy should be sent to the Moderators at their private email
address.

Thank you.


-HW "Skip" Weldon
Columbia, SC

Thanks Tad.

On Feb 23, 4:06 pm, beliav...[at]aol.com wrote:

- quote -

FWIW I enjoy your obvious deep knowledge, insight and analysis.

I've sent you a personal email on this topic (from darkness39 (at)
yahoo (dot) com). You might find it illuminating.

On Feb 20, 5:02 am, sawyervill...[at]gmail.com wrote:
- quote -

Optimization", described as follows:

"Portfolio Optimization uses modern portfolio theory to identify
efficient combinations of stocks and mutual funds with the minimum
level of risk for a specified annual expected portfolio return. It
includes a built-in database with annual returns for the past 10 years
for all stocks in AAII's Stock Investor Pro and all of the mutual
funds in AAII's Quarterly Mutual Fund update. It also plots a graph of
the efficient frontier."

I just played with the software for a few minutes and have some
reservations.

(1) It is a DOS program, which is ok with me, but it crashes when one
specifies a date rate of 2000-2007, since it has data only up to 2006.
That's not how to write a program.
(2) It handles a maximum of 32 securities or mutual funds.
(3) It uses only yearly data.

I am thinking about using portfolio optimization of mutual funds and
ETFs for my own investments. The first hurdle is getting accurate
(adjusted for splits and distributions) daily return data. One can get
daily data for stocks and funds from Yahoo Finance using the Python
programming language using a script at http://rimonbarr.com/repository/pyq/index.html

"John Gunn" <noway[at]forgetit.org> wrote
- quote -

Do you think fraudulent advertising did take place here?
Because otherwise, your comments are a non sequitur to my
own.

Both of you need to take your objections up with the poster
who acknowledged his original comments were inaccurate, re
appearing to predict the future. Your determination to
ignore this point and try to chill the marketplace of ideas
is striking.

"Elle" <honda.lioness[at]nospam.earthlink.net> wrote in news:juNEh.4345
$PL.2666[at]newsread4.news.pas.earthlink.net:

- quote -

Sadly, Tad is right on. Your posts are ill-informed on the reality of
investmenting and now securites regulation.

"Tad Borek" <borekfm[at]pacbell.net> wrote
- quote -

Nothing stated in this thread constitutes fraudulent
advertising. I am sorry you object to my pointing out an
inaccuracy in a statement on portfolio optimization. Doing
so goes towards educating the public on financial
forecasting, just as pointing out that the many mutual fund
managers who ostensibly "beat the market" in fact according
to studies mostly do not in the future.

The inaccuracy which I pointed out was acknowledged by its
poster. He went on to elaborate that his actual literature
on his software is more precise. No harm, no foul, on Usenet
nor anywhere else. I do enjoy your posts. They're very
revealing. :-)

Elle wrote:
- quote -

Hmmm, now an expert on securities law. Let describe the reality so you
understand our point of view.

Any written communication, including usenet and bulletin board postings,
can be considered "advertising" under state and federal securities law.
This isn't my opinion -- an auditor told me this. There are very
specific guidelines regarding advertising, and as you might imagine,
"fraud" is a very dirty word in that context.

You may think this isn't the case, or think it's a bad law, but it's the
way it is. This is why we see no registered representatives
(stockbrokers) posting on MIFP, their broker-dealers generally forbid
it. Us independents can decide for ourselves, and believe me, the
compliance attorneys say "don't take the risk!"

The question isn't whether posting here could cause problems, or whether
one reader could cause problems for us -- it's why any of us would even
bother, once there's a risk.

-Tad

"Tad Borek" <borekfm[at]pacbell.net> wrote
- quote -

For the record, IMO Usenet posts hold virtually no water in
a court of law as far as "fraud," inter alia, is concerned.
The reason being that this is part of the free marketplace
of ideas. Like all of Usenet, this is a casual forum where
people are not under a legal obligation for perfection in
writing. I would not dream of reporting, using legal
avenues, anyone here. Such a complaint would make a
laughingstock of the complainer. I am surprised you do not
know this.

I am sorry you have such a problem with dissent in a forum
like this. It is in fact vital to the success of commerce
and industry.

I fully support diversity in one's portfolio. I think the
aforementioned portfolio optimization tools (for fee or not)
are far more likely to do good than bad. My objection
remains that their output should not be presented as being
definitely superior in the future.

"Tad Borek" <borekfm[at]pacbell.net> wrote
- quote -

See above. That's not what I said.

Elle wrote:
- quote -


Don't underestimate how much this kind of nonsense -- accusing a
professional poster of fraud in connection with securities -- shuts down
contributions to this newsgroup. At least, contributions from those of
us who actually have access to tools like portfolio optimization software.

I wouldn't dream of talking about what I use, or how I use it, lest my
comments not meet Elle's capricious standards for disclosure,
non-solicitation, and heck grammar. One phone call to a regulator by a
loose-cannon MIFP reader and I'm potentially looking at a one-week
non-routine audit. Why would I bother?

Kastnna, I'm sure plenty of people appreciate your posts...but from
someone else in the biz...consider yourself warned. <sigh
-Tad

<beliavsky[at]aol.com> wrote
- quote -

I agree.

Re selling stuff: It's about generating buzz (and IMO not a
few urban legends within that buzz) as much as plugging
one's specific service.

On Feb 23, 3:10 pm, Will Trice <wwtr...[at]paragondynamics.com> wrote:
- quote -

Since volatilities are the square roots of the diagonal elements of
the covariance matrix, the simplest approach is to use a single window
and sampling interval to compute the covariance matrix. A good
introduction to estimation methods is Chapter 16 of the book "Modern
Investment Management" by Goldman Sachs Assset Management.

One way to study the goodness of correlation/covariance estimates is
to look at the returns achieved by an optimizer using them as inputs.
The paper below (available by purchase only) tried computing
covariances using 5 years of monthly data and 1 year of daily data.
For long-only portfolios there was not much difference, but for long/
short portfolios daily data looked better.

Minimum-Variance Portfolios in the U.S. Equity Market
Clarke, Roger de Silva, Harindra Thorley, Steven
THE JOURNAL OF PORTFOLIO MANAGEMENT
Fall 2006
In the minimum-variance portfolio, far to the left on the efficient
frontier, security weights are independent of expected security
returns. Portfolios can be constructed using only the estimated
security covariance matrix, without reference to equilibrium expected
or actively forecasted returns. Empirical results illustrate the
practical value of large-scale numerical optimizations using return-
based covariance matrix estimation methodologies, providing new
perspective on the factor characteristics of low-volatility
portfolios. Optimizations that go back to 1968 reveal that the long-
only minimum-variance portfolio has about three-fourths the realized
risk of the capitalization-weighted market portfolio, with higher
average returns.

A working paper by Tal Schwartz at
http://www.departments.bucknell.edu/...27Schwartz.pdf
found good results using 1 year of daily data. The good results may
largely be the result of the 1-year momentum effect, since he
estimates mean returns of stocks using the last year, which I would
not recommend.

These studies look at optimizing a portfolio of stocks rather than
asset classes. Both impose limits on the amount invested in any one
stock, 5% in the latter case.

One can Google "beliavsky realized volatility" in this newsgroup to
find a paper on asset allocation using portfolio optimization and
implemented using futures contracts -- not practical for most people.

In article <1172500877.202041.50410[at]v33g2000cwv.googlegroups.com> ,
"kastnna" <kastnna[at]auburnalum.org> wrote:

- quote -

Sure; that's entirely reasonable. My suggestion was that Bernstein,
despite some hyperbolic criticism of MVO tools in his book, supports
this same kind of approach -- his website, after all, is named
"efficient frontier", and stuff on it seems to me to imply that his
consultancy operations work analogously to what you describe above.

On Feb 26, 12:07 pm, "Elle" <honda.lion...[at]nospam.earthlink.netwrote:

<snip
- quote -

Even if an ETF has only been around for three years, it may have an
underlying index with a much longer history and an underlying asset
class with an even longer history. Most ETFs have tracked their
indices pretty closely -- this has been studied. I am sure the
software uses index and asset class data for its historical studies.
As usual you don't know what you are talking about.

If kastnna is trying to sell his products on this newsgroup, he is not
trying very hard. I don't know what his real name is, and there is
nothing in his signature mentioning his company (there would be
nothing wrong with mentioning this). Instead of trying to censor
others, censor yourself!

"kastnna" <kastnna[at]auburnalum.org> wrote
- quote -

You don't even need the disclaimer. You need only state that
this is what history has shown.

- quote -

The very basis for _illegal fraud_ is xyz, "intent" can be
nebulous, it depends on the state, and so forth. My meaning
was meant more casually.

You're welcome in advance, because I think there is no doubt
that one's sales pitch is enhanced by speaking as though
past performance represents the future. It's so easily
corrected, grammatically, and it's so important, that I
personally find no excuse for not using the right language.

'specially when you're talking about, what, three years of
ETF data? People have to make a living, and I do support
diversification, but this newsgroup is about a lot more than
people selling their products.

On Feb 26, 9:35 am, "Elle" <honda.lion...[at]nospam.earthlink.net> wrote:
- quote -

I don't intend to mislead the noobs and I don't think they can be to
any serious extent. Even if they become convinced that I (or someone
else) "have all the answers" they are never going to not run into the
necessary disclaimers at some point in the process. If they somehow
managed to track me down, the first thing they would hear would be
"you know there is no guarantee of future performance..." I don't end
every verbal sentence with a disclaimer either when talking with
clients, but before we make any meaningful progress the disclaimers
are well understood.

Long story short its caveat emptor, and "you get what you pay for." If
you like you can attach a disclaimer for me everytime I post. I'll say
"thanks" now in advance.

FYI, by definition fraud cannot be "unconscious". The very basis for
fraud is intent. Careful tossin' that word around.

On Feb 26, 9:01 am, beliav...[at]aol.com wrote:
- quote -

Somewhat. We do "look back" to compare performance after the
portfolios are constructed. The software is only used for fee-based
clients. These clients get quarterly reviews that cover performance,
rebalancing, AND the performance of their old portfolio had they kept
it. It reassures clients to see what would have happened had they not
changed their portfolio. The company that developed the software,
fiserv, does most of the legwork for us. We get a wealth of
performance data against numerous benchmarks. Among others, actual
return and standard dev. are given so that we can do sharpe ratios.

We do not perform sharpes on ALL clients because many had portfolios
that were so poorly constructed there is no question. Many
underperformed us AND had higher Std Dev.

The clients that had fairly well constructed portfolios to begin with
tend to be our "more demanding" clients by virtue of being more
educated. We usually do sharpe ratio values with them during their
review. We are confident so far, but remember we have only been using
the software for three years and many of the ETFs we use aren't very
old either.

"kastnna" <kastnna[at]auburnalum.org> wrote
- quote -

Words are everything. Newbies overwhelmingly frequent this
group.

Just my opinion, but it serves the "financial critical
thinking skills" of America etc. well to add an extra
qualifying sentence (or words) to posts like your previous
one. Otherwise, one could make a perfectly reasonable
argument that there's some attempt at fraud (unconscious or
otherwise) and seduction here when one uses phrases like
"Our company's allocation is more efficient than yours."

On Feb 26, 9:02 am, "Elle" <honda.lion...[at]nospam.earthlink.net> wrote:
Sorry Elle. We attach our lovely little "based on historical data"
disclaimers on everything necessary in our office. For the sake of
brevity I leave it off on a non-regulated usegroup.

If the readers here honestly believe that I can accurately predict the
future and/or have software that can assure outpreformance then they
have much bigger problems than can be addressed in this group.