Who else does custom indicators? | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

Custom Indicators vs. Technical Indicators 6 replies

does anyone else not close losing trades? 14 replies

VTTrader: How to program custom indicators at CMS?? 4 replies

Custom indicators 4 replies

Does anyone else feel like the Fed is blowing smoke? 12 replies

Attachments: Who else does custom indicators?
Exit Attachments

Who else does custom indicators?

  • #1
  • First Post: Jan 25, 2007 5:21am Jan 25, 2007 5:21am
My group and I cough up a lot of custom indicators, that fit our requirements.. I am wondering if any other person or groups are cranking out custom indicators, and more specifically, are they superior to the indicators out there already?

I'm more curious really, about how good "we" are against those who have put out famous indicators..

Although I'm not allowed to disclose the indicators themselves, I've attached a screenshot:

Green and Red dots indicator 2nd order derivative inflection points

Red and Blue lines mathematical mobile support and resistance stop lines

Red and Green Histogram indicates trend directionality

Curious how others are generating custom indicators to solve what specific problems.
Attached Image
google:
  • #2
  • Jan 25, 2007 6:45am Jan 25, 2007 6:45am
I did try to create one a while back that dealt with derivatives of a moving average but just tossed the idea aside.

andersenws
"Youth is the trustee of prosperity." - Benjamin Disraeli
  • #3
  • Jan 25, 2007 7:40am Jan 25, 2007 7:40am
I have one of the first order derivative also, but because the moving average "meanders" along, the first order derivative was not as useful as the 2nd order, in which we can see market sentament change.. Bulls vs Bears, who has the upper hand, is clearly defined in the concavity shift...
google:
  • #4
  • Jan 25, 2007 8:49am Jan 25, 2007 8:49am
My concavity shifts every few minutes otherwise I'll would wind up with a aching back at the end of the trading day.
  • #5
  • Jan 25, 2007 10:04am Jan 25, 2007 10:04am
Once I developed an indicator based on Momentum. I backtested and forward tested it for a while and it looked like the Holey Grail but then it started doing funny things. Make it short, I realized that it was good for some market conditions only so I put it aside.
  • #6
  • Jan 26, 2007 6:22pm Jan 26, 2007 6:22pm
I'm working on a custom indicator to calculate slope of the Money Flow Index using a linear regression of the last (n) periods... I'll probably post it here when it's done.

On a related note--what do you all think is the best method to calculate the derivatives of an indicator? 2-point slope? Multi-point linear regression? Moving average of either of those? Something else?
  • #7
  • Jan 26, 2007 7:10pm Jan 26, 2007 7:10pm
I'm definitly indicator happy, as you might guess from my nick. Here's a small sample of indicators I've done. Paint bars and profit lines (show profits and losses for trades) for MACD Momentum, Clock and Grid and a Woodie's CCI suite with CZI and Sidewinder.

I have a lot more then I'm not showing here.
Attached Image
  • #8
  • Jan 29, 2007 10:21pm Jan 29, 2007 10:21pm
Quoting brue
I'm working on a custom indicator to calculate slope of the Money Flow Index using a linear regression of the last (n) periods... I'll probably post it here when it's done.

On a related note--what do you all think is the best method to calculate the derivatives of an indicator? 2-point slope? Multi-point linear regression? Moving average of either of those? Something else?
Something else.
  • #9
  • Apr 22, 2007 12:29pm Apr 22, 2007 12:29pm
Sorry I missed this thread; I've just put one based on LSQ fitting over on the MetaTrader forum. The chart below is one sort of similar to what I'm seeing above - it has three lines - one to estimate the part of a trend coming from jumps (unusual events) - in yellow, and the part coming from from average size variations - in red. The blue line is the AGM of the other two. I call it PrimaDonnnas vs BeanCounters (or PrimaBeans for short). I use it mostly to just check behaviour patterns on the $CAD, which is prone to jumps alternating with flat movement, in a trend. I can put up the code (which is as is, and very unpolished), if anyone asks, but right now I think there's better ways to approach this.....

http://gammarat.com/Forex/30070420/usdcadpb.gif
  • #10
  • Apr 22, 2007 12:43pm Apr 22, 2007 12:43pm
Quoting brue
I'm working on a custom indicator to calculate slope of the Money Flow Index using a linear regression of the last (n) periods... I'll probably post it here when it's done.

On a related note--what do you all think is the best method to calculate the derivatives of an indicator? 2-point slope? Multi-point linear regression? Moving average of either of those? Something else?
Probably a Kalman Filter (see the wikipedia artical for a simple version). I'm programming one up ATM for price and price velocity, but it will be a while before it's done, as it's a bit complicated programming it up in Metatrader (which is what I'm using), which doesn't have a matrix library. Otherwise linear regression on a line (in the derivative you are looking at using differencing), and take the leading point, or variations thereof (e.g. running estimates of mean and variance), weighting estimates according to prior and post variance.
  • #11
  • Apr 22, 2007 2:06pm Apr 22, 2007 2:06pm
Quoting twoblink
My group and I cough up a lot of custom indicators, that fit our requirements.. I am wondering if any other person or groups are cranking out custom indicators, and more specifically, are they superior to the indicators out there already?

I'm more curious really, about how good "we" are against those who have put out famous indicators..

Although I'm not allowed to disclose the indicators themselves, I've attached a screenshot:

Green and Red dots indicator 2nd order derivative inflection points

Red and Blue lines mathematical mobile support and resistance stop lines

Red and Green Histogram indicates trend directionality

Curious how others are generating custom indicators to solve what specific problems.
Yes, inflection points can be nice to detect accelerating movements, but unfortunately they will confuse you much of the time on volatile pairs because strong sudden movements are usually followed by profit taking, and worse: the profit taking is anticipated so you get strong opposing movements right after. In other words, you will get a lot of false positives.

The histogram indicator looks rather decent though. Looks more complex than a MA. Please do post both indicators though. Thanks.
Virtue finds and chooses the mean. Aristotle, Ethica Nichomachea
  • #12
  • Apr 22, 2007 2:47pm Apr 22, 2007 2:47pm
[quote=twoblink;198059]My group and I cough up a lot of custom indicators, that fit our requirements.. I am wondering if any other person or groups are cranking out custom indicators, and more specifically, are they superior to the indicators out there already?

I'm more curious really, about how good "we" are against those who have put out famous indicators..
=========================================

LOL -- let me say this about that !

forex, as compared to equities, is a phenominally simple market to live in -- it obeys laws of support and resistance and works with an amazing "bumper car" mentality, in which it hits support and bounces to resistance ! How long this takes depends totally on the timeframe one is trading, as it happens intraday and at discernable times of the day, over a 10 day chart period and for a longer time period, on the daily charts !

I myself have designed two "indicator systems" that are presently in constant use and has reached over 7000 downloads, BUT they are based on the timetested older indicators which EVERY pro trader is using and consist of nothing more complex than the LRC for the timeframe one is working in, and a few of the very standard indicators, arranged to show present movement and trend reversals. The greatest indicators for forex are support and resistance, the Demark high and low pivot points and the fibs --- using them becomes a "self fullfilling prophecy" because ALL banks and pro traders are using them ! Oh yeah, theres a stochastic, a MACD, an RSI and a CCI thrown in there also, but way up there, above your wildest guess on any currency, is a stop set by the banks on their longer term options dealings, and that stop was generated by the fibs !

its fun to watch people re-invent the wheel (take a look at the more recent murry math lines compared to the 7 centuries old fibs, or whatever is trying to replace the bollingers at the present time --- check out price channels and gasp at how well they predict !)

But the pros still use the old reliables, which is HOW they became "old reliables" in the first place AND why these traders are the PROS !

enjoy and trade well
mp
  • #13
  • Apr 22, 2007 2:53pm Apr 22, 2007 2:53pm
Ahh, time was when pilots used to refuse to fly by wire, and before that knights that thought that arrows were for sissies. Remember Agincourt
  • #14
  • Apr 22, 2007 3:02pm Apr 22, 2007 3:02pm
Quoting HalifaxCB
Ahh, time was when pilots used to refuse to fly by wire, and before that knights that thought that arrows were for sissies. Remember Agincourt
===================================

LOL -- technology is one thing, but oh so different from what the truth of trading is all about.

when a better indicator is found, the pros will be using it -- but visit a pro shop NOW and check out what they use --- ask a barclays or dean whitter trader what they use !

NEVER say never, but for RIGHT NOW, and for the past 50 years or so (600 for the fibs) these indicators are what work, are being used and where the BIG profits come from !

When trading one can deal with the past, the present or the future --- all i am graced with knowing is the present and of that i speak ! (hey, the future has many twists and turns and theres "many a slip twixt cup and lip"

just a simple trader, using simple indicators in a simple way --- sounds pretty simple to me !

enjoy and trade well

mp
  • #15
  • Apr 22, 2007 3:25pm Apr 22, 2007 3:25pm
mp - I agree in general, and still use traditional methods a lot (mostly Fibo's, RSI, and MACD). OTOH how many times do we hear "the market doesn't trade like it used to".

But with a strong background in tracking and track analysis (mostly for ASW applications) + a lot of work in optimization, I have to say that most indicators are rather sluggish, so I rather enjoy transferring techniques that might (say) guard a ship against a high speed evasive torpedo (where you've got maybe 30 seconds to make a life or death decision) to market analysis. Sometimes, the results are surprising (like the LSQ estimator I put over on the MT forum) and sometimes they are goofy (like the one I put above). But as big money turns increasingly towards quant analysis (and anyone who feels it isn't should check out the options market, which wouldn't exist without it), I figure it's time for small traders to put a new edge on the old weapons...
glgt..
CB.
  • #16
  • Apr 23, 2007 3:34pm Apr 23, 2007 3:34pm
Just a little question - just how old are the "old reliable" indicators? Wasn't Wilder (fwiw, an engineer by training) doing his work in the 70's (his big book was published in 78) and 80's, and when were fib levels actually introduced to the markets? The cncept has been around for centuries, although the application is how old? Appel (MACD) is still around, so we can reasonably guess his method isn't all that ancient...AFAIK, the oldest western chart techniques are the P&F - which is still only a little over 100 years old.....
  • #17
  • Apr 24, 2007 3:16am Apr 24, 2007 3:16am
Quoting HalifaxCB
Just a little question - just how old are the "old reliable" indicators? Wasn't Wilder (fwiw, an engineer by training) doing his work in the 70's (his big book was published in 78) and 80's, and when were fib levels actually introduced to the markets? The cncept has been around for centuries, although the application is how old? Appel (MACD) is still around, so we can reasonably guess his method isn't all that ancient...AFAIK, the oldest western chart techniques are the P&F - which is still only a little over 100 years old.....
================================================

Well, lets see here -- is not the classic definition of an antique 50 years ?

Wonder how many we have approaching that exhalted status ?

C'mon now and loosen up 30 or 40 pips -- the truth of the matter is that the "standards" are the ones being used presently and for a number of years now (longer than both you and I have been trading, and I've been trading a long time !)

A floor trader will go into the commodities market armed with nothing more than a list of the days pivot points and an overhead view of the upcoming bids and asks (level 2) and come away at the end of the day with bonus making trades --- she/he isnt even using indicators at all.

Market makers and specialists have a bit more time and can take a fast look at a rising stochastic or RSI, and on a 1/4 min or less chart, make a fortune.

Now thats talking about intraday timeframes, and when one is talking about longer term "position" or "swing" trades, things change slightly. Now we get into "waves" and the half art/half science that entails, although I usually hear practitioners state something along the lines of "present resistance area MAY be the beginning of wave 3"

Now few are those who can predict the future with unerring accuracy, but a lot of the indicators come pretty close when used with any knowledge of how they work. I, once for a contest and once for the fun of it and published for 7000 people to read, decided to predict the price of two stocks extended out to the future. I would have won the contest, but that someone else had already taken the price i wanted and the second time had people laughing because i missed the price of the stocks run up . . . . . . . .

BY $ 0.02

Was I great --- nah --- i used the same old tools that the traders would be using in actual trading of the two equities, and by using the same tools arrived at the same place as they did, but ONLY because we were all working on the same page with the same tools !

When the "new" tools become the "old" tools, Ill use them also and throw away the "antiques" --- but until and when that happens, I will work with my old fashioned stuff, cause everyone else is old fashioned also !

imo, my thoughts and my money goes where my mouth is

enjoy and trade well

mp
  • #18
  • Last Post: Apr 24, 2007 5:29pm Apr 24, 2007 5:29pm
Quoting HalifaxCB
mp - I agree in general, and still use traditional methods a lot (mostly Fibo's, RSI, and MACD). OTOH how many times do we hear "the market doesn't trade like it used to".

But with a strong background in tracking and track analysis (mostly for ASW applications) + a lot of work in optimization, I have to say that most indicators are rather sluggish.
==================================================

interesting but my answer is yes, no, maybe !

an indicators speed is highly variable, dependent on the values inputted vs the amount of "whipping" one can accept, and while most indicators can be "smoothed" one of the greatest concepts to keep in mind is TIMEFRAME !

the longer the timeframe, the slower and less whippy the indicator. For intraday trading, i change trend direction based on a 1,4,10 MACD which also extends to 2,5,10, but would never advise those values for a 6 month daily chart, where the WADI, the WTSI, the ADX/CCI swing pointer and even the venerable momentum and ATR stand out as direction and strength indicators.

I speak of the "old reliables" only because the rest of the trading world uses them, and understand that there was probably much resistance in using them when they first appeared on the scene, but the truth is that i trade NOW, and what may be in the future can easily change what i will be using, but for NOW, this second, this hour, this day --- the old reliables are what makes people decide one way or another, and ill stick with how they have performed and their place in the pantheon of tools.

one note --- all of the new ideas seek to create THE SAME RESULT as the oldtimers, but in a new way, which is why i refer to it as re-inventing the wheel. By way of example, if wave theory tries to determine reversal points and the effect of those reversal points on a charts movements, its simply applying a "new" science to one established by support and resistance areas, the LRC's and the fibs. Its end result is to predict, and every generation feels its "latest and greatest" is better than the old. (which may or may not be true, but appears to be true to the adherents of the new method)

I shall wait and see, but the original question in this thread was "how are the new ones doing against the old" and my present answer, just like my past answer, is that theyre just not up to snuff just yet !

perhaps soon, but not yet !

my opinion, but it seems the majority of traders HAVE NOT yet switched !

enjoy and trade well

mp
0 traders viewing now