|
PureBytes Links
Trading Reference Links
|
Bill, you can't input your own indicators into ICE. You can test your
indicators with the systems tester like always but you will have to
do the programming yourself.
The 74 indicators that ICE tests are the ones in Technical Analysis A
to Z, and come as a subset of MS.
Those indicators are the most commonly used ones. I've reviewed and
tested somewhere over 500 indicators (the actual number is close to
700 but I've lost count somewhere in the blur) None of the ones I've
tested are really any better than the ones that come with MS. It's
just that different people have different personal perferences.
As far as the thresholds go, the indicators are first tested with the
thersholds that are commonly used such as the RSI(14) with 70 30
thresholds. When the indicator is optimized on the group of stocks
you select, the period is optimized along with the thresholds. So the
report gives you all the information that's needed--after
optimization you might get an RSI(22) thresholds 75 25 for example.
StochOSC and Stoch Momentum Index are the only Stochs included. They
are the ones that come with MS to begin with.
The bottom line is you can write your own code for whatever you want
to test in the Systems Tester. ICE gives you the parameters it uses
to run systems tests, so you can setup your tests on your custom
indicators the same way as ICE. ICE simply has consistent and stable
tests and optimizations already programmed for the 74 indicators that
are included.
I know you'll probably disagree with this, but I've looked at dozens
of SToch indicators, and none of them tell you anything that can't be
read from simpler standard indicators. Almost everybody in TA,
especially the ones who write books, feels the need to come up with
new versions of indicators so their stuff looks unique. It sells the
books. But in reality, it's hardly unique. Out of every 500 or so new
indicators that get programmed and written about, there's one that
makes something work better, but that's about it.
Here's a link to e-analytics that explains it in detail. Their
article summarizes my experiences in looking at indicators.
http://www.e-analytics.com/sysind.htm
For me, this program was well worth the price. $99 is nothing in the
world of TA.
JO
--- In Metastockusers@xxxxxxxxxxxxxxx, "Bill Hebert"
<billhebert@xxxx> wrote:
> JO, Got a few questions?
>
> I like the concept of ICE but would like to hear from an existing
user
> if it is also flexible enough for users to add their own indicators
to the
> the program or change how the indicators are being used in the
program.
>
> Can the ICE program accept new indicators in the 4 different groups
> that you state for managing colinearlity?
>
> For example: If there was no StochRSI in the Velocity indicator
section
>
> Could I add my own and have it work with the program?
>
> Or if there was a StochRSI, could I add my own version to the
program?
>
> Another question is on the usage of indicators
>
> Can the ICE program allow for unique indicator usage?
>
> For example: The RSI uses say 30 and 70 for trigger lines.
>
> Can I set to trigger when it goes above or set it to trigger when
it falls
> back below that same level. Goes above 80 to trigger or drops
> down below 80 to trigger?
>
> Or say I want to use the RSI with a 50 trigger line crossover
instead
> of a dual 30/70 line overbought/oversould crossover indicator.
>
> I have reviewed that ICE program from the site but did not see
anything
> that would support answering my questions.
>
> Thanks,
>
> -- Bill
>
>
> ----- Original Message -----
> From: "manohohman" <kelols@xxxx>
> To: <Metastockusers@xxxxxxxxxxxxxxx>
> Sent: Wednesday, October 22, 2003 5:30 PM
> Subject: [Metastockusers] New Metastock Add-on Reviewed
>
>
> I just got an early release copy of a new MS add-on called ICE.
>
> Before reviewing the product, here's a disclaimer for all the
> conspiracy theory compulsives out there. I have no involvement with
> either Equis or Adaptick, the company who produces the program. I'm
> not married to the sister of someone who works for either one of
> them, and nobody there is my cousin. They don't send me birthday
> cards with money inside, and so far neither of company has even
> bought me a hot dog.
>
> The product's name is ICE, which tells you absolutely nothing about
> the program. I will say this: I REALLY LIKE THIS ONE!
>
> I haven't said that often about any add-on but this time, I really
> got my money's worth out of it.
>
> ICE is a specialized system tester program that takes 72 indicators
> and divides them into four relatively uncorrelated categories. The
> indicators in each category can be tested on, and optimized for,
any
> group of stocks in any time frame.
>
> What this product does is save huge amounts of time and energy when
> building and testing trading systems.
>
> All trading systems rely on some kind of indicators. However, if
you
> are going use two, three or four indicators, each indicator should
be
> measuring something different. For example, if you like RSI and
> Stochastics, fine. Just don't put both of them in your system
because
> you are essentially measuring the same thing with both indicators.
>
> Both are momentum indicators. RSI measures closing price relative
to
> the previous close, and Stoch measures closing price relative to
the
> range. This might sound different, but they're too closely
correlated
> to be very effective when used together.
>
> In the ICE program both of these are considered velocity
indicators.
>
> The point of having different indicators is to increase the
> probability of a good trade by increasing the amount of NEW
> information provided by each indicator. If the indicators are too
> similar-measuring essentially the same thing-the combination only
> marginally increases the probability of making a better trade. If
the
> indicators are sufficiently different, then the new information
makes
> a larger and more important contribution to the increase in
> probability of a good entry or exit point.
>
> However, that's not the end of the story. Even if you have
indicators
> that have very little colinearity (okay, take deep breathes-it
means
> they're measuring the same things-i.e. Points on a straight line.)
> there is still the issue of how well any given indicator works on a
> group of stocks.
>
> In a good trading system all the indicators give good signals on
the
> stocks being traded, and they are distinct enough to provide sound
> entries and exits when used together.
>
> (I need to pause for a moment to make a small point outside of ICE.
> With indicators, more is not merrier. The best trading systems keep
> it simple. As you add more and more indicators to a system, the law
> of diminishing returns sets in and JO's law of paralysis takes
over.
> If you use more then 3 or 4 indicators, it gets progressively
harder
> to make decisions based on what you're seeing. Half the time you
> won't even know what you're seeing, much less what to do about it.
So
> basically---KEEP IT SIMPLE!!!!)
>
> After installation, ICE is fully integrated with the systems
tester.
> You pick a couple of indicators in a category, test the indicators
on
> your stocks to see which ones perform the best, and then optimize
> them. The optimizations are also fully integrated, so you don't
have
> to do any programming at all.
>
> Once you have the optimized indicators, you set them up on chart
with
> a stock and play around a bit with the settings. This how I like to
> customize. My eyeball tells me things that a mechanical
optimization
> program can't even deal with. Once that's complete, you move to the
> next indicator in the next category.
>
> After awhile you will have three or four indicators, one from each
> category, that fit you, your stocks and your system. They will
have
> little colinearity, they will have excellent probability of
> generating good entry and exit points and you will be well on your
> way to having a really good trading system.
>
> No, the program doesn't test chaos theory, Elliot cycles, fractals,
> or hypocycloids of four cusps. This is a good thing because it has
> been proven over and over, the best systems use the simple time
> tested indicators--and that's it. Two moving averages and some
common
> sense will out perform the stuff I just mentioned. Okay, don't
email
> me about how good Elliot cycles are and how much you love them,
etc,
> etc. Elliot was stranger than Elvis, and Elvis didn't know anything
> about trading stocks. That's all I need to know about Elliot and
> Gann.
>
> This program won't help you write rap music, or find exotic
patterns
> in the futures markets, but it will save you tons of time and money
> trying to come with a system that works.
>
> Here's a real life example-mine. I have written and developed two
> really, really good systems. (No, I won't give them to you. No, I
> won't sell them to you. Don't bother!) But like all Americans (and
> the Chinese too), I'm never happy with what I have, so I wanted a
> third one that did something that my first two didn't. My original
> systems gave very accurate entry and exits signals but they didn't
> tell when I should be piling more money into what looked like a
good
> trend. While it's not possible to tell how long a trend will last,
it
> is definitely possible to tell when a particular trend is stronger
or
> weaker. In other words, when should I risk more or risk less.
>
> I started working on the system a couple of weeks ago, and went
about
> solving the problem of when to increase my trade risk first. After
> that was solved, the indicators that I added initially to the
system
> seemed to perform only so, so. After getting ICE, I read the manual
> (which is still a work in progress, but since none of you seem to
> read manuals anyway that shouldn't be a problem), sat down at the
> computer and in less than two hours finished the system. I've set
it
> up, traded with it, and I'm pleased to say---I LOVE THIS SYSTEM, I
> REALLY LOVE THIS SYSTEM. Next week I'm going to buy it a three
stone
> diamond ring, I love it so much. Now, if it could only tell me it
> loved me back!
>
> If I had had this product a few years ago, it would have saved me
> hundreds of hours in system development and testing time. That was
> such grinding work, I had days when I thought I saw Elvis-sorry I
> mean Gann.
>
> You know the only thing that could be better than this is if I
could
> tell you it was FREE, FREE, FREE! That is, after all, your favorite
> word. Alias, it's not. BUT ALMOST.
>
> If you buy it before the official release date, it's only $99
instead
> of $299. Well, hold my hand, I can't believe it. It's true, and I
> will tell you I would have paid many times that amount for all the
> time and energy it would have saved me when I first started
> developing my own systems.
>
> You can read about it at www.adpatick.com.
>
> I guess before I finish I need to confess. No I'm not John Slauson,
> the guy who wrote ICE. Yeah, I know JO is really similar to JS-at
> least to some of you Sherlock Holmes want-ta-be's. Given the
millions
> of possible name combinations JO and JS aren't even remotely
> related.
>
> My confession is about cycles. I do rely on some cycle information.
> You can find it at
http://www.fourpillars.net/finance/predic.html.
> This is the most accurate cycle information I've ever seen. If you
> don't believe me, use it for awhile. You'll see. I've named my dog
> Yang.
>
> JO
>
> PS Some of you know I recommend John Clayburg's book Four Steps to
> Trading Success. John has developed a very sophisticated,
effective,
> yet easy to use set of indicators based on the RSI, Stoch and %R.
It
> would seem that these have too much colinearity to be effective.
Doc
> Clayburg has solved that problem through the use of time frames.
Time
> frame combinations can over come colinearity problems, but it's too
> sophisticated for most system developers to deal with so I won't go
> there, except to say Clayburg's system does not suffer from the
> problems I have described above.
>
>
>
>
>
>
> To unsubscribe from this group, send an email to:
> Metastockusers-unsubscribe@xxxxxxxxxxx
>
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Rent DVDs from home.
Over 14,500 titles. Free Shipping
& No Late Fees. Try Netflix for FREE!
http://us.click.yahoo.com/mk9osC/hP.FAA/3jkFAA/zMEolB/TM
---------------------------------------------------------------------~->
To unsubscribe from this group, send an email to:
Metastockusers-unsubscribe@xxxxxxxxxxx
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|