Tom:

I feel gratified that you deem it a good read.
I can't use the last 3 of the ratios mentioned when
looking at a spreadsheet & the one about Liquidity
isn't findable either. The CurrentRatio and
TotalDebt/NetWorth are immediately useful though.
Both WorldCom AND Enron had CurrentRatios that
were in the 1.0 plus or minus .1 for several years
before they finally collapsed. When I was evaluating
companies to get interested in, I passed over them
without thinking anything else about it.

No company should owe more than it's worth, but
if it does, probably has a problem servicing its
debt, so TotalDebt/NetWorth is easy to find and
relevant to a company's well-being and profitablity.

As to Liquidity, because I can't find AccountsRecievable
on the quarterly spreadsheets of corporations, I substitute
Cash/CurrentLiabilities. By this meassurement, I'm
surprised AOL can pay their monthly lights and utilities,
nevermind their stock.

From the Enron/WorldCom debacles I've decided that
the statistics concerned with shares can be manipulated
too easily. I've found something that works pretty well
and intend to stick with it. I'm sure, you'll probably
be just as happy with the Costales way of looking at
financial information.

As always,

Warren_

Mr Leapyear92:

I'm going the way I'm going, doing what I'm doing
regardless of Tom's IW. No one can read the
future, and I doubt Tom can either, but I do
have hopes he's right.

Hoping for the best,

Warren_

Conrad:

I read the information you sent about VORTEX.
I really liked the geometric series you generated
on page 11 and how you used it on BuyOrders and
PortfolioControl. I stumbled across the exact same
relationship. I also saw that under some circumstances
you equate PortfolioControl and Equity. I don't
do this in a program, but I've found it useful
to do this exact thing on occasion.

I'm most of the way into a starting position for
a first stock and needed to know how much I needed
before I'm ready to start trading FOR REAL.
found out how much I have left to go by setting
PortfolioControl & my StockValue(equity) equal then
claiming my average cost of my shares was the proper
share price today. On the next line I put in the
REAL stock price and my version of AIM told me
I needed to buy so many shares more before we could
start working the AIM system FOR REAL. God I hope
Tom's IdiotWave is accurate. If so, the bottom is
near.

The "Lichello Flaw" irked me too but, I found
that if you use the SAFE boundaries to flag a
permission to trade then buy or sell the whole
value of Advice, the "Lichello Flaw" _goes_away_.
I never saw you subtact out a SAFE value
like in StdAIM (-AlsoKnownAs- AIM(0)) so I'm
betting we buy we do this the same too.

I've also calculated two SAFEs in such a way
as you can take the first one, add 10% to it
and you have PortfolioValue. If you add 10% to
that you have the second SAFE.

SellSAFE-PortfolioControl > PortfolioControl - BuySAFE

The two SAFEs constitute what you describe
as a "deadband".

I've heard a lot of noise about SplitSAFEs
and the difference between prices on the way
up and prices on the way down before I saw
your take on that "problem". Actually, the
relationships work out nicely if you look
at it as a problem involving logarithms. If
you express the price as a logarithm and figure
out what a 10% difference in price would be
in terms of logarithms, the 10% up is the
same as 10% down. I related the two SAFEs
logarithmicly. It's not the normal way of
doing things, but that's what makes us tinkerers
tick isn't it.

Your compounding the gains of minimal investing
periods is one way of motivating a daily
regimin(sp?) on users. I agree that a daily
regimin(sp?) is really better, but I've
mentioned Nyquist before.

You included a lot of factors of one sort or
another I wouldn't have thought to include,
but I tinkered with a number used in buying
that you talked about some on page 8. Lichello
divided the amount of shares bought by 2
before feeding this number back into PortfolioControl.
I thought the number 2 was not holy and my
method seems to work very well if we use
34/3 instead of 2. I wouldn't suggest everyone
use that number, but it works on my system.

Anyway, I read the material you sent and was taken
back that two people working independently could
arrive at such similar results.

You've included some things I wouldn't have
thought about including, but the similarities
of conclusions are frightening.

As ever,

Warren_

Conrad:

I understand. Make SAFE wide enough so that whatever
trades happen are profitable. That's basicly what
this is about. Again, I'm a small investor and
trading with paper more than real money (for the
moment) and have investigated where the maximum
profit happens when I alter SAFE and then, keeping that
constant, reduced the number of shares/dollars until
the profit became zero, adusted upward a little, then
rechecked where the maximum gain was again by adjusting
SAFE and again, holding that constant, reduced the
principal until the profit moved to inside of
$5 over the time frame of interest. This, using
a set of REAL stock numbers of a stock said to
be trading in a sideways manner. I've also done
this with just a waveform.

I believe I have a grasp on the minimum investment
necessary before profits overtake expenses, but
I'll investigating this with a more widely
varying SAFE and a minimum tarde size A LOT MORE.

I wish you well,

Warren_

Conrad:

As I mentioned before, I was basicly ignorant
of all the discussions that went on here before
I arrived, so I only have reference to Lichello's
work and the work of Mr Webber (who mostly echoes
Mr Lichello's work). Yes, I'm referring to
AIM(revision 0), the original, when I say AIM(0).
I'm not too sure about how to do subscript zero
here. Anyway, I don't remember AIM(0) referencing
a minimum trade order so it must be something
that's sprung up here.

Mr Lichello did mention a SAFE value of zero and
that if a stock was sold the second it made a
penny over cost, that that was no way to make
money, nor was it a good idea to buy at the
first penny of a discount. The difference
between buy and sell prices would produce
$.02 profit per share; that's no good.
Mr Lichello discusses this point exactly on
page 42, paragraph 3 & 4 of his book's third
revision. In the following paragraphs he comes
up with SAFE, but what's said in the paragraphs
I mentioned is identical to having SAFE=0.

As I said, I know nothing about this MinimumTrade
and so, I defer to you gentlemen to tell/show
me something about it.

Although I can't see how to use it yet, please
tell me more,

Warren_

Qarel:

You make no profit with SAFE = 0, and I don't think a
a "minimum order size" will work well. I haven't tried
it yet so I don't know quite how to go about "making
it happen" so I'd need to listen to the chat/debate
before I'd have any clear idea about how to go about
implementing it.

Of course I'm interested in trading the volatility.
Mr Lichello argued that the algorithm NEEDS
volatile stocks to work well. I've heard of X_DEV
and as I understand it, it uses "moving averages"
and the AIM algorithm to determine how much to
buy and when. If my guess is right, you need to
update the X_DEV software daily so that it can
develop the "moving averages". My solution
tells the owner at what price the next buy (or sell)
will happen. All the operator has to do then is
wait until the StockPrice crosses those marks.
Once the mark is crossed, enter the price, read
how many shares to trade note the next
Buy/Sell prices and wait or go golfing (or whatever)
until the StockPrice crosses those new marks.

I worked out a lot of bugs (I think) and its operation
has been made supremely simple.

As for just purchasing X_DEV, I pretty much had
my spreadsheet created and finished before I got
here and heard about X_DEV for the first time.
From what little I've seen of most of the
AIMish software, it seems the BuySAFE and
SellSAFE numbers are almost always 5 or
10%. The people who operate this seem to be limited
about what numbers they can pick from; that's a frowny
face. I don't know, limited selections and having
to daily log in to tell the software the date and
price of the shares I'm interested in is a small
chain, but a chain nonetheless. If I had it handed
to me in a box, I'd probably start tinkering shortly
after opening it.

Engineers are all tinkerers at heart it seems.
Should I be embarrased?

Oh well, still

I wish you well,

Warren_

Conrad:

I'm waiting. I gotta see this!

Warren_

Conrad:

As for thinking things out, I sometimes forget
to "count the cost" and pay for it. I can get your
address off this bulletin board, but I stated mine
in post #270.

I'm not an Aquarius, just careful.

Anyway, hi ho, hi ho, it's off to work I go.

I wish you well, and good morning.

Warren_

Conrad:

I was doing something else, remembered something
you mentioned and thought I had to respond.

You were right, I do calculate gain per trade(sorta).
I calculate gain per BuySell pair. No money is
made until something is sold. I sampled five
waveshapes:
(1) square wave, amplitude=1.0, offset=10,
(2) sine wave, amplidute=1.0, offset=10,
(3) square wave, amplitude=.5, offset=10,
(4) first 36 prices, NYSE-NDX, Lichello(rev3),pg106,
& (5) one month of JDSU, 8.8.02-9.8.02.
I figured the gains, less gains in moving averages
(where applicable) and took the (number of trades/2)th
root. This comes out as Gain/TradePair but varied some
from sample to sample so I multiplied 5 rates together
and took the 5th root. Preliminary results indicate
a gain of 1.45% per TradePair. You get to figure out
how many Buy/Sell TradePairs have to occur for an
investment to double, then estimate how long it
takes to do this.

Yes, I figured it out as Gain/Trade. I bet you did
something similar and wanted to know if my results
were similar to yours. I hope I've answered the
question.

As always, hoping for the best,

Warren_

Qarel (or is it Karel?):

Thank you for the information about GTC.

I'm aware of AIM(revision 0), SAFE and
how it reduces AIM(0)'s MarketOrder size.
As a small investor, I have to trade
more shares than AIM(0) authorizes just
to break even. Once I recognized this basic
problem, it didn't take me long to figured
out that SAFE should be used as a
"permission to trade" and no more. Once
permission to trade has been given, the
question becomes, "how much may I prudently
trade?". This is one of the easiest places
where one can say AIM(0) needs optimization.
I've found one sort of optimization, but
should someone find a better answer, I'm
not so proud that I won't switch methods.
Until that time I believe the VORTEX method
is extremely similar to anything I've found
and deserves examination.

I think that covers it, so for now
I wish you well,

Warren_

Conrad:

I'll second the barbeque.

For the time being, send an email to
lw_rogers@sbcglobal.net and I'll send you what
I've got. I'll happily examine your VORTEX and
from what we've exchanged so far there's a good
chance we'll end up marveling in the similarities
and reveling in the differences much like two
chess players enjoy each others company.

Again, like chess players congregating on a
chess bulletin board, I'll probably visit
the VORTEX section of this bulletin board,
much more frequently.

For the present, I wish you well,

Warren_

Conrad:

I believe we have quite possibly arrived at points that
are indistinguishable and if your method is copyrighted
or patented, I shall say no more.

Yes we should meet and maybe throw a barbeque together!
Actually I'm not that good a cook except for a mean
potato salad, but you get my drift.

As for the SAFE being different from AIM(0), it turns
out that with my development of SAFE, you can take
the BuySAFE and multiply it my (1+x%) and end up with
PortfolioControl, and multiply this value by (1+x%) again
and end up with SellSAFE. This these numbers are related
to each other by a CONSTANT PERCENTAGE and are actually
more linearly related than the SAFE Buy/Sell numbers
mentioned by Mr Lichello. What is interesting is
that if you divide the BuySAFE and SellSAFE numbers
by the number of shares you own, two interesting numbers
pop out. You get (1) your next target BuyPricePerShare
and (2) your next target SellPricePerShare.
At this point you need only see if today's SharePrices
exceed these numbers. You only need do something
if these numbers are exceeded, otherwise do nothing;
today's sampling and updating of AIM has been done.

As to bias, there is also a "reverse biased diode".
Biasing, in so far as I've seen it, merely means
something like "making something (a transistor
or diode) useful by setting the voltages and
currents (by way of a circuit arrangement) to
predetermined (useful) levels".

Actually, the x in the "1+x" term is a percentage.
Multiplying "a variable" times (1+x) is the normal
way I've always added x percent to "a variable".
There's nothing special there.

As to an "F" being some freely defined AIM function,
really we must chat about that some time. I'm
late to work and your making more of it than it
really is. If you saw it you'd say, "is that it?".

Hoping for the best,

Warren_

Conrad:

I forgot to comment on making graphics.

I'm not sure about the presentation of computer
graphics. The last time I had anything to do
with graphics really was when Pascal was the
language "de jour". I'd have to ask my wife
(BSCS summer 2002) to get me updated about
HTML, presentations & such.

Originally I starded with a PortfolioControl,
and a split SAFE featuring PortfolioControl/(1+x%)
& PortfolioControl*(1+x%). This seemed to work
fairly well, but I just had to know what my
real costs were and I wasn't happy when AIM
sometimes suggested I should sell shares below
what they REALLY cost me. This led me to throw
in another term that wouldn't allow sales
suggestions unless they were GREATER THAN my
costs per share + 10%.

Also, SAFE is more like a "bang-bang" controller
in that SAFE is a boundary, that when crossed,
"wakes up" the controller, realigns the controlled
elements (setting the controlled system back on course)
& resumes the "sleep" mode. Crossing the SAFE
boundaries should normally give permission for
other things to happen, but otherwise plays no
other part in a control system's response. In the
AIM variant I've worked on, SAFE gives permission to
trade whatever "Buy/Sell Advice" wants the system
to trade once the SAFE boundaries are crossed.
I know this isn't the standard way to develop SAFE
or even SAFE's canonical use, but it seems more
like something one might run across in a Controls
Systems journal than before. Hope no applecarts
have been overturned.

Hoping for the best,

Warren_

Conrad:

You lost me.

As I understand it, at any one moment the price of
a stock is equal to (1) a static (or long term
moving average) term and (2) a dynamic (or very
short time period) term. If the static term is
truly a constant, the B&H investor never profits
either on paper or in terms of cash. Both the AIM
investor and daytraders are hoping to capitalize
(to one extent or another) on the dynamic term.

Where I work, sine wave generators, have a knob for
waveform type, a knob for frequency, a knob for
waveform amplitude (or size), a selector switch
that allows one to insert an offset bias & a knob
that controls the size of the offset one chooses
to insert. In other words;
the output voltage at any instant of time =
the offset bias (static [or constant] term) +
a sine (dynamic term) wave.

To be honest, I probably should never have used the
word "bias" because that's probably not used the
same way outside my workplace as inside it.
I really meant no confusion.

I only use two cycles and end the period of concern
as I started it, at $10. This insistence on starting
and ending the period of interest at $10 (or any
given constant) is to clarify that the B&H investor
makes zero profit, but profit is still possible for the
AIM investor because of AIM's relationship to a stock
price's dynamic term. This is almost identical to a
discussion whose spreadsheets supporting the argument
is found between pages 64 and 71 of the 3rd revision
of Mr Lichello's book. I just restricted Mr Lichello's
argument to a somewhat more narrow range and I use
the well defined waveform to optimize the relationships
among the numbers in my spreadsheet. This is very
similar to an electronics technician isolating circuit
problems via sine wave insertion.

I hope this helps clear up any confusion I've
somehow inserted in this discussion as well as
clarify that my use of a sine wave/fictitioius
stock price didn't originate with me.

Hoping for the best,

Warren_

Conrad:

I feel really pleased that someone that
really knows what's going on thinks the
AIM variant I've cobbled together is
worthwhile.

As for "bias", I should probably have
used "change in bias" (from electronics).
If you buy a stock at $10 and it has a
"change in bias" of $5, the ending price
(for the period of concern) is $15. The
Buy&Hold investor gains $5 per share.
Imagine a stock that had a constant bias
of $10, but varies only enough to trigger
an AIM authorized buy/sell pair, the AIM
trader should show a profit while the
Buy&Hold investor's profit is zero.

Yes, you got my point about cycles, missed
opportunities & knowiing where the next trade
levels are although I'm not sure what a
"GTC Order" is. I'm really a newby at this,
could you enlighten me?

Warren_

Tom:

I seem to have totally forgotten the social graces.
I wish you a Happy Birthday and may the fortunes
of you and your household be good ones.

Sincerely,

2Penny

Tom:

You know, your IW is as close to a crystal ball as I know
of. I've almost recommended it to others elsewhere as a
reason to cheer up.

I'm not altogether happy being a newby during this time,
but OTOH given that my one account (so far) doesn't go
belly up, I'm just about invested enough that my broker
won't be eating up my profits while I execute AIM directed
trades. The other enemy is stock price deflation. I hope
we're at the end of this wringer.

Did you ever get a look at that Costales book?

Well, enough shop talk, for now.

I wish you a good evening.

Warren_

Conrad:

I've been working on a version of AIM. The Nyquist
criterion states that sampling MUST occur at least
twice as fast as the fastest frequency you are observing
to accurately reconstruct the wave. For AIMers, more
profits could be possible by more frequent use of the
AIM algorithm. The problem with that proposition is
AIM's tendency to eat up cash in a sideways market.
Ok what set of values do you feed a "test case AIM"
so that you can see and adjust for this tendency?
Personally I use a fictitious stock price of $10
with a slowly varying sinewave of amplitude 1.0
($10 +/- 10%). After a few cycles(no bias), I measure
the gain and check for other anomalies.

I used square waves for a while, but gave a definite
gain while SAFE never got a chance to work.
I believe the sine wave simulation of the stock market
will provide the best simulation without that
pesky bias (characterized by a Buy&Hold person making
a profit). A final question about using a sine wave
is size and frequency. Much is made of 10% in many places
so I also used that for the sine wave's size and
I used a slow frequency corresponding to 2 cycles/month.

Conrad, like you, I can't look at things and leave them
alone, I've got to tinker.

Warren_

Hi Tom & Everyone else:

I understand about things that eat up your time.
There are two times to do any particular thing;
(1) by schedule, (2) when necessary. As I understand
it AIM should be operated on a schedule. I was
working on a method based more on necessity.
If I knew at what price the next trade should
happen, all I have to do is wait for it and
check the DOW, Nazdaq, or whatever daily.
When the stock I'm interested in crosses the
buy or sell boundary I do something. Gee, I
might even have it all prearranged with my
broker and when the trade happens I figure out
what prices and quantities should happen next,
arrange it with my broker and go on a hunting
or fishing trip until then.

So how is everyone here holding up? With the
stock market diving as it has been, I'd like
the lowdown on how you all are riding it out.

Until mine gets bigger though, I'll continue
getting to work on time.

Warren_

Hi All:

Tom said I should get in contact with the people here.
I'm also working on a flavor of AIM. It spits out
what the next Buy/Sell prices are. I keep track of
the current stockThe stock moves and crosses one of
the two barriers and once crossed, I check the
spreadsheet and buy or sell a quantity based on
the latest price and quantity specified by AIM.

On first blush it seems like it'll be a winner, but
then again I'm not sure. Trade for trade AIM(0)
[AIM(subscript zero) - the original] is generally
better, but this doesn't trade by a schedule. A
lower gain might be tolerable if more frequent
trading makes up the difference.

That doesn't cover it all, but it's good for
a bit of an introduction.

Warren_