Australian (ASX) Stock Market Forum

Quest for a NEW predictive BUY signal indicator

Joined
8 January 2009
Posts
48
Reactions
0
Have you noticed that most of the technical indicators are very old, most being developed more than 20 years ago, and many 50 years ago, well before the age of computers.

There are two issues here: 1. computers provide a fundamental shift in the way indicators can be developed. 2. Computers have fundamentally changed the way the entire market functions and behaves because computers provide instant information.

If you look at the details of the derivation of the indicators that all of us use you will see that most are very primitive and simplistic, despite their widespread use.

I have been looking for new modern indicators that provide better predictive power. My idea is to focus on parameters that change just before a stock starts a new surge. Sure there are many sudden announcements etc. that are unpredictable. But often there will be signals in the few days prior to a rise that are partially due to the way that the market behaves and the effect of some new activity or interest in the stock. I have been focusing on the following parameters used to provide a buy signal:

Increase/decrease in Volume
Close in the top half of the daily range
The size of the daily range
Gaps i.e. sudden increase in mean daily price compared with yesterday’s mean.

I have been developing an index that uses a combination of these parameters. So far the results look promising and the index gives a signal one to four days before the price surges.

I would be interested on your comments about these and other signal parameters and new indicator methods that attempt to do the same thing.
 
Let me save you alot of time right here and right now.

You will never find the holy grail.

Brad
 
lemme know how you go! im always up for any insights, back to the track for me, jokings

jc
 
I have a bit of time on my hands today now that lunch has been cancelled so I will bore you all to tears with a couple of trades.
Mainly this is for the benefit of those looking for 'that' magic indicator.

The price behaviour is really all that matters, indicators and volume etc are an indication of what may happen next with the price pattern.

Trade 1...
Below are two pics of EHL in Metastock, an interesting chart pattern that has potential. I want to find stocks with that type of pattern, how ?
In the past I have looked at these patterns and then spent a lot of time placing various indicators on the chart and then modifying the indicators so that I can create a scan that may pick up these patterns.

The chart pattern is what is important, the indicators can be used as a way of finding a chart pattern, that's what I am doing in this case.

The first chart is my scan result and the second chart displays all the modified indicators that result in the green arrows on the first chart and consequently the stock appears in the search.
(The last arrow, last night's, was a crossing of the Siroc)
There are numerous other inputs to the scan (vol etc) to eliminate the chaff.

I hope these are of some help in explaining that indicators are just a tool and used in conjunction with a pattern may result in some $$$.

(click to expand)
 

Attachments

  • EHL Ind 020409 2.png
    EHL Ind 020409 2.png
    26.1 KB · Views: 16
  • EHL Ind 020409.png
    EHL Ind 020409.png
    34.7 KB · Views: 11
Following on from my previous post.

Trade 2...
This again is pattern based using Elliott Wave instead of indicators.

In this case we have a potential 5 wave sequence playing out and if so then we should be nearing the end of wave 4.
The first chart shows that W.4 is in the right area.
The second chart shows the potential target area if a W.5 plays out and the third chart shows the entry (mine) and stop positions.

The volume that you now buy is based on the size of your capital and the percentage of it that you are going to risk between the entry and the stop if this pattern fails, and they do fail more often than reaching their target.

Thats where risk/reward and trailing the stop up to breakeven minimises the loss and allows you fight another day... any other way is gambling in my opinion.

(click to expand)
 

Attachments

  • OSH 020409 1.png
    OSH 020409 1.png
    21.7 KB · Views: 8
  • OSH 020409 2.png
    OSH 020409 2.png
    20.4 KB · Views: 5
  • OSH 020409 3.png
    OSH 020409 3.png
    25.1 KB · Views: 7
i have found its all how you look at things.

what part of the story does an indicator tell?

what other indicator(s) could complete the story?

i dont think the wheel need re-inventing. just a new way to look at something.
 
i have found its all how you look at things.

what part of the story does an indicator tell?

what other indicator(s) could complete the story?

i dont think the wheel need re-inventing. just a new way to look at something.

I use a suite of indicators to screen/scan for likely stocks to invest in.
I use a suite of indicators to closely examine the selected stocks for likely buy and sell signals, as well as the price candle chart etc. This issue is about whether there is scope for new modern indicators that may be more atune to the 21st century behaviour and patterns of the stock market. Though I know little about VSA ( I'm learning), what I am suggesting is an indicator based on similar principles - volume, price range, position of the close and gaps. The idea is that these parameters are likely to change in the days prior to a surge and that these signs can be built into an indicator as a buy signal. This is a new round wheel. I have recently looked at the formulae for the indicators I have been relying on and was quite shocked at their age and how primitive they are ->>> Model T Fords. Computers have changed the market behaviour and trading patterns but we are still reliant on pre-computer indicators and methods.
 
i have found its all how you look at things.

what part of the story does an indicator tell?

what other indicator(s) could complete the story?

:iagree: ---- pretty much sums it up, although U and I are looking at FX, which may be a bit more conducive to typical T/A Norm

I use a suite of indicators to screen/scan for likely stocks to invest in.
I use a suite of indicators to closely examine the selected stocks for likely buy and sell signals, as well as the price candle chart etc.

What do u mean by closely Rus? --- do you use a lower time frame with your scanning for entries---- ie Hourly vs EOD etc ? --- just curious.


. Though I know little about VSA ( I'm learning), what I am suggesting is an indicator based on similar principles - volume, price range, position of the close and gaps. The idea is that these parameters are likely to change in the days prior to a surge and that these signs can be built into an indicator as a buy signal. This is a new round wheel.

stocks are not my forte, but VSA coupled with price action/position seems very valid, because its 'current' information, at any time frame -- i speak with no experience in that regard though (Tech -- others ?)


. I have recently looked at the formulae for the indicators I have been relying on and was quite shocked at their age and how primitive they are ->>> Model T Fords. Computers have changed the market behaviour and trading patterns but we are still reliant on pre-computer indicators and methods.

has human behaviour changed over the last few years? ---- a "primitive" indicator seems perfect for a primitive beast :D

Momentum may be the only indicator which will do what you are looking for Rus -----

Why? because Momentum will be dropping often b4 price does (or rising etc etc) ----- ie when Momentum diverges or "leads" price --- that is the only 'indication' you need (in my humble opinion) ------ cheers.
 
Thanks for your comments cartman, I would like to comment briefly on

Momentum may be the only indicator which will do what you are looking for Rus -----

I use MACD etc. BUT it is sobering to realise that


Momentum is simply


Closing Price [today] - Closing Price [n days ago] That’s it!!!!!

It is a following parameter that exaggerates the trend over a set period of time.

It is critically dependent on the value of n - 5 days, 7 days, 10 days, 21 days.

Example

Closing Price - Closing price [5 days ago]
5 day Momentum Indicator

[Closing price CP]
1 52.00 - -
2 51.00 - -
3 51.50 - -
4 48.50 - -
5 53.00 - ...CP 5 days AGO....momentum
6 53.50 .......52.00 ..............1.50
7 53.50 .......51.00 ..............2.50
8 54.00 .......51.50 ..............2.50
9 54.00 ........48.50 ..............5.50 (spike error)
10 55.00 .......53.00 ..............2.00

Perhaps our ignorance is bliss and better maybe we don't know or understand fully about the derivation of the tools we use.

I for one was ignorant about their simplicity. Knowing more worries me - that I have been misusing the signals as trends likely to continue.

Cheers,
 
Thanks for your comments cartman, I would like to comment briefly on

Momentum may be the only indicator which will do what you are looking for Rus -----

I use MACD etc. BUT it is sobering to realise that


Momentum is simply


Closing Price [today] - Closing Price [n days ago] That’s it!!!!!


G'day Rus ---- ps i like the way U think BTW -----

When i say Momentum --- i am not strictly speaking of the "Indicator" Momentum --- if u get my drift ;) ---

Momentum can be calculated by "Indicators", thats true, ---- but that would be too easy wouldnt it :D ---

Its the quest of "calculating" the 'important' Momentum that will give you the answer to your questions ---------

TH does it with his understanding of DOM for the Index ----- ( way (repeat --- WAY) out of my league , but considering im only trading bucket shops, im happy with my progress ---------- i dont profess to know anything special, but Momentum relative to Time relative to the instrument you are trading is all you need to know -----

if i was trading stocks, id only trade 2 or 3 and concentrate on their behaviour ------ cause you would get to see how the 'smart money' works them over -----

in essence, that is all trading is about for us small guys ---- work out how the big guys are operating ---- whether it be through VSA, E/W, Momentum, or whatever ------ the big boys run the market --- the little guys have to learn how to work out where the big guys are heading (at a given point of time)

ps im on my meds, so take it with a grain of salt ;)
 
Computers have changed the market behaviour and trading patterns but we are still reliant on pre-computer indicators and methods.
what are some typical market behaviour changes due to computers? Maybe look at the cascading stop runs which are a consequence of electronic trading.
 
Perhaps our ignorance is bliss and better maybe we don't know or understand fully about the derivation of the tools we use.

I for one was ignorant about their simplicity. Knowing more worries me - that I have been misusing the signals as trends likely to continue.

Rusty most good traders who have been trading for a while understand what these indicators are and how they are constructed. And that's why very few actually use them.
Some of the elite traders like FrankD construct thier own indicators(if you can call his pivot points an indicator) and system, I think Nick Radge also uses some custom indicators to determine divergence but reliance on these indicators in minimal at best.

I'm not sure what stage you are at but IMO you are better off developing a complete system and method using an arm of T/A or mixture of them - essentially learn to read a chart. The system also needs to include trade management & money management methods, entry & exit rules etc, etc.

You will find there is no such thing as a magic indicator and you will chase your tail trying to build the perfect indicator forever. Smarter people then you & me have tried;)

Good luck

why complicate things?

Exactly.
 
why complicate things?

AS my Gold Coast Mate used to say:

You can't stop progress

The aim is to simplify changes in the 4 parameters mentioned into a signal.
This is simpler than looking at the chart and examining how they are changing and their interaction.

Also the aim is to use the indicator as a stock scanner/filter which is simpler.

So why not simply 4 parameters into a single one with enhanced power.


Cheers,
 
Top