Opinions wanted on a new sector fund investment website

If you're interested in sector fund investing with Fidelity sector
funds, I'd appreciate some feedback on a commercial website I'm
developing.

The goal of the site is to make it relatively easy to manage a sector
fund portfolio with only a moderate amount (averaging 10 or less) of
exchanges per year. The basic approach is a momentum strategy, but one
that is enhanced through use of some additional statistical/numerical
methods to minimize the amount of transitions (ie "whipsawing") during
flat market periods.

One thing I'm doing that's different (as far as I can tell!) is that I
have our back-testing system "online", so you can test performance over
different periods, and then graphically see the trading results, along
with a history of every purchase or exchange made during this period.
In other words, you can pick a period of time that you think would make
a good test, and see how our back-tested results would look. I don't
know anyone else that does that.

You can sign up and use it for free for 60 days. After 60 days, you'll
have the option to continue as a subscriber if you want. But if you
provide some good feedback, I'll be glad to give you a free
subscription for a year. At this point, I'm just looking for good
objective feedback, not necessarily subscribers.

As an FYI, the following articles might be of interest to anyone
interested in sector fund investing and if momentum strategies have
some documented evidence of performance.

Hulbert's Review on Sector Newsletters


An academic study on industry sectors and how they affect stock
momentum

Re: Opinions wanted on a new sector fund investment website

Sorry, the link got clipped off the end of the message!

The site url is

Re: Opinions wanted on a new sector fund investment website

<> wrote in message
news:
> Sorry, the link got clipped off the end of the message!
>
> The site url is
>

That's OK, we have Bob.


.... and he doesn't charge after 60 days!

Re: Opinions wanted on a new sector fund investment website

In article <Abn2e.26662$>,
"Gary C" <> wrote:

> <> wrote in message
> news:
> > Sorry, the link got clipped off the end of the message!
> >
> > The site url is
> >
>
> That's OK, we have Bob.
>
>
> ... and he doesn't charge after 60 days!

hey Gary,
this summer I'm going back up to Ely. got to find Bob's address in MN.,
leave a case of something good on his porch. I owe Bob.

Re: Opinions wanted on a new sector fund investment website

I love how Bob doesn't charge after 60 days, it's definitely the best
perquisite of his deal for sure! Thanks to Bob!

Re: Opinions wanted on a new sector fund investment website

I see Bob has some vocal supporters! I checked out his site.
Interesting, but I'm doing something quite different. And free is nice,
but that's a minor consideration if you're investing any sizable amount
of dollars. A few percent increase in annual return can easily pay a
subscription fee many times over.

TopSectors gives very clear buy/hold/sell signals for a model portfolio
of three sector funds as well as ranking all the Fidelity sector funds
based on trends (but using some different metrics). There's no
ambiguity on what to do, which makes it very clear what TopSectors
performance is over time. TopSectors also tracks your sector fund
holdings and emails you when it's time to switch.

But the biggest question is which system gives the best overall return
over a specific time period. I can't answer that for his system so I
don't see how he can optimize a trading strategy. For example, how do
you know that the optimal value for some switching metric is X versus Y
if you can't simulate and backtest with those different values and see
which has a better risk-adjusted return? Or what's the optimal size for
your sector portfolio- 1/3/5/10 or X number of sector funds? That's NOT
saying he hasn't done so, but at least with a quick review, I can't see
how. Of course, backtesting can't guarantee future results, but it at
least gives a way to quantify performance of different systems over the
same time period. Unlike anyone else, I let you backtest over ANY time
period from 01/01/95 to present, so it's impossible for me to tweak
numbers to look good over one or two fixed time periods.

For example, take the time period of 1/01/00 thru 12/31/02. Equity
indices had a significant drop. But backtesting our system over that
time returned 16%. You can run the same backtest by entering this link:

(it may take a minute or so for the page to load as we're getting a lot
of people running simulations). And you can change the start and end
dates to a different time period if you want.

So if you're getting better results with another approach, great! I'm
just curious how you're doing it, in a way that I can verify and test
against.

Mike

Re: Opinions wanted on a new sector fund investment website

<> wrote

>I see Bob has some vocal supporters! I checked out his site.
> Interesting, but I'm doing something quite different. And free is nice,
> but that's a minor consideration if you're investing any sizable amount
> of dollars. A few percent increase in annual return can easily pay a
> subscription fee many times over.

And that 'few percent increase' is gauranteed how?

> TopSectors gives very clear buy/hold/sell signals for a model portfolio
> of three sector funds as well as ranking all the Fidelity sector funds
> based on trends (but using some different metrics). There's no
> ambiguity on what to do, which makes it very clear what TopSectors
> performance is over time.

'Over time' as in the future. Good one.

There must have been hundreds of ads just like this one posted here over the
years and they all claim to have the answer. Truth is, they don't.

Remaining sales pitch deleted out of kindness.

Re: Opinions wanted on a new sector fund investment website

>And that 'few percent increase' is gauranteed how?

My point was the cost of a subscription is easily recovered if you get
marginally better performance. If you subscribe and don't feel that
way, you get your money back. That's what TopSectors guarantees.

>'Over time' as in the future. Good one.

My point was that it's very clear what TopSectors recommends, so it's
easy to compare against other methods. If method A is specific about
what to buy/sell/hold, and method B is open to interpretation, then you
obviously can't compare the two methods without making assumptions
about how to implement method B. You're correct, that isn't as
important now as it will be in the future, other than from an
ease-of-use standpoint. But it also indicates that TopSectors isn't
hiding behind ambiguous recommendations.

>There must have been hundreds of ads just like this one posted here
over the
>years and they all claim to have the answer. Truth is, they don't.

There are some significant differences:

1) No one can guarantee performance, but I can say that, statistically
speaking, sector-momentum newsletters have outperformed other
approaches in the past. I enclosed a few links in the first post, one
by Hulbert, another an academic study, that supports momentum
strategies. If you invest in ANYTHING, you're (presumably) making a
prediction that it is going to perform better than the alternatives.
How else would you rationally do that, if not based on
statistically-supported approaches?

2) Can you name anyone who's put their backtesting system "online" for
public use, for any user-defined time period? Backtesting is not
perfect, but it certainly beats smoke and mirrors or wild guessing! If
a system shows better performance than non-managed indices over a set
of random time periods (which is how TopSectors trading strategy is
optimized) I would at least give it more credibility than a system that
provides no such
validation.

So I'm curious - what is your investing philosophy? Why do you think
this strategy would be less effective than the one you currently use?

At this point, I'm as interested in getting valid feedback as anything
else - hence the offer to GIVE 1 year subscriptions to people who
provide useful feedback. I know I have to qualify that, so I'll say the
first 20. That's hardly a get-rich-quick scheme.

Mike

Re: Opinions wanted on a new sector fund investment website

<> wrote

>>And that 'few percent increase' is gauranteed how?
>
> My point was the cost of a subscription is easily recovered if you get
> marginally better performance.

Key word here is "if". No gaurantee.

If you subscribe and don't feel that
> way, you get your money back. That's what TopSectors guarantees.

Ok, so I follow your advice, lose $10,000, you give me back my fee.
I'm all tingley.

> 1) No one can guarantee performance, but I can say that, statistically
> speaking, sector-momentum newsletters have outperformed other
> approaches in the past.

Most sytems will outperform from time to time. They also underperform from
time to time.
For any reasonable length of time just about all of these newsletters fail
to beat their benchmark.

> 2) Can you name anyone who's put their backtesting system "online" for
> public use, for any user-defined time period?

Lot's of them do. The problem here is that I can create a wonderful trading
HISTORY too. The real problem is that you can't go back in time and invest
in the past.

> So I'm curious - what is your investing philosophy? Why do you think
> this strategy would be less effective than the one you currently use?

I don't think I said anything that would indicate that I have a system that
would beat some historical data that could have been cherry picked.

> At this point, I'm as interested in getting valid feedback as anything
> else - hence the offer to GIVE 1 year subscriptions to people who
> provide useful feedback. I know I have to qualify that, so I'll say the
> first 20. That's hardly a get-rich-quick scheme.

Mike, nobody accused you of running a get rich quick scheme.
You might be very sincere, the problem is that nearly all advisory services
aren't worth the money, it's really that simple. People would generally do
much better by investing the subscription fee.

If you want to set yourself apart from the others....

<> wrote

> You can sign up and use it for free for 60 days.

......post your trades here, in advance, for 60 days. I'm sure that if you
have a good 60 days then you'll get lots of subscribers.

No one has ever been willing to do this. It would take you almost no time to
post your picks here and people could see just how great your system is.
Anyone unwilling to do this just doesn't have enough confidence in his
system to interest me.

Re: If you want to set yourself apart from the others....

Good point! No, I have no problem with that, and that's a smart way to
document predictions over time. Once a week, for two months, I'll post
recommendations to this discussion for so it doesn't clutter the
newsgroup with new discussion topics and tick people off. I'll start a
running tally of cumulative percent return starting next week.

TopSectors Picks for 3/31/05

FSENX FSNGX FSESX

Mike

Re: If you want to set yourself apart from the others....

You are this first to do this and I admire your courage. Good luck with it.

I've set up a $10,000 portfolio and bought 87 shares of FSENX, 68 FSESX, 97
ESNGX.
I have $29.63 left. Purchases made at the close of business 3/31/05. When
posting a new fund be sure to tell us which to sell to free up the cash for
the new purchase. I'll also keep track of the $7.50 exchange fee and the
0.75% short term trading fee (30 days).




<> wrote

> Good point! No, I have no problem with that, and that's a smart way to
> document predictions over time. Once a week, for two months, I'll post
> recommendations to this discussion for so it doesn't clutter the
> newsgroup with new discussion topics and tick people off. I'll start a
> running tally of cumulative percent return starting next week.
>
> TopSectors Picks for 3/31/05
>
> FSENX FSNGX FSESX
>
> Mike
>
>

Re: If you want to set yourself apart from the others....

Ed wrote:
>
> I've set up a $10,000 portfolio and bought 87 shares of FSENX, 68
> FSESX, 97 ESNGX.
> [...]
> I'll also keep track of the $7.50 exchange fee and the
> 0.75% short term trading fee (30 days).

"[Y]ou will not be charged an exchange fee if you exchange through any
of Fidelity's automated exchange services."

Select Portfolios prospectus.


--
Mark Freeland

Re: If you want to set yourself apart from the others....

Thanks for the reminder, I should have known that but I haven't used Selects
for awhile now.


"Mark Freeland" <> wrote in message
news:
> Ed wrote:
>>
>> I've set up a $10,000 portfolio and bought 87 shares of FSENX, 68
>> FSESX, 97 ESNGX.
>> [...]
>> I'll also keep track of the $7.50 exchange fee and the
>> 0.75% short term trading fee (30 days).
>
> "[Y]ou will not be charged an exchange fee if you exchange through any
> of Fidelity's automated exchange services."
>
> Select Portfolios prospectus.
>
>
> --
> Mark Freeland
>

Re: Opinions wanted on a new sector fund investment website

Of course you have optimized your system using past market results.
Then developed a strategy for duplicating those results based on future
market movements. If the future market behaves like the past market,
your system may work. If not, it won't.

Re: If you want to set yourself apart from the others....

Thanks, but this wasn't really courage. True courage is investing your
IRA and your spouse's trust fund. ;-)

I appreciate your taking the time to test how well this works. Sixty
days is tough to verify an equity investment approach - but you have to
draw a cut line somewhere. I'd bet that this trial will be determined
by how the energy sector does. Also, you pretty much did the
all-or-nothing method. I think it would be interesting to backtest that
versus a staged approach - ie, invest 10% from cash every 2 weeks. I
may test that idea as well and see what happens. ...

FYI.. if you'll register, you'll get an email when it's time to trade.
That's closer to an actual user trial that having me post to usenet.
I'll still post to usenet, since that documents and timestamps for me
by default, but it might be worth your trying as well, since TopSectors
crunches the numbers nightly and sends out email notices
automatically. I'd expect an email once maybe once a month right now..

I appreciate the feedback I've gotten so far. I figured this would be a
tough audience, which is what I was looking for. There's one thing that
I think you misunderstood - I agree, most people post their backtested
results as "proof". If you are trying to be deceptive, you can easily
pick some period and tweak your trading to be optimal for that period.
So in addition to some history, I've put the backtesting SYSTEM
online-not just history. So you can enter any period you want and see
what happens. I think that would be a lot harder to fake. It also
takes into account trading penalties if you exchange <30 days -
something that bothered me about the other system someone mentioned
since he said he ignored fees when backtesting. Seems like a major
thing to ignore.

Someone else raised a good point about optimizing over one period, so
as long as history replays itself according to that period, you're
fine. The main difference is that I optimized over many periods (ie
good, bad, 2 year, 6 month, etc) for the best average result. Not
rocket-science, and still prone to historical basis, of course, but I
think less so. I can't prove thats true, but it seems to make sense to
me intuitively. Shoot me down if you think that's off-base or you think
there's a better way.

Mike

Re: If you want to set yourself apart from the others....

<> wrote

> I appreciate your taking the time to test how well this works. Sixty
> days is tough to verify an equity investment approach - but you have to
> draw a cut line somewhere. I'd bet that this trial will be determined
> by how the energy sector does. Also, you pretty much did the
> all-or-nothing method. I think it would be interesting to backtest that
> versus a staged approach - ie, invest 10% from cash every 2 weeks. I
> may test that idea as well and see what happens. ...
>
> FYI.. if you'll register, you'll get an email when it's time to trade.

You can do that here as well, your choice.

So far I'm down $30.76 using your picks.

Re: If you want to set yourself apart from the others....

Time to sell FSNGX. Trade for FSPHX... Use the closing prices as of
April 12.

Ed, I'd like to make sure we're calculating return the same way - I had
different results from your post on April 5. Can you email (or post)
how you are calculating return?

Since FSNGX got hammered 8% in one day, I'd imagine it's down about 2%
as of tomorrow...