Developing a mechanical system from scratch

[CanOz]

Heres the bars.

 

[lesm]

Does anyone here have intraday data and tesing capability.
If so i'm sure you'll find that (For stocks) time of day entry is not a biggie.
Futures---different.

Yes, but on another platform, not AB.

Only have ASX EOD and Forex intraday in AB.

Happy,

As a separate discussion, I would be interested to see if your are finding are finding any real "edge" in trading the 5th hour on the ASX. Wouldn't take long to test it out.

Using particular trading window (time frames) in futures or Forex can make a difference and not an unusual approach.

 

[happytrader]

Hi Tech, Lesm

No need for anyone to waste time or use anything but free data to test this observation. www.bigcharts.com covers the last 10 days trading in short time frames. Just print or copy them to file at the end of each week. You need to invest your own time and to prove it to yourself if you are intending to trade with real cash. I did this for 2 years on selected blue chips as well as watched the ticker tape until I was satisfied that I could use this to gain an edge in my trading.

If you want to actually observe traders trading the last volatile half hour of the market then you can go to www.paltalk.com

Cheers
Happytrader

 

[doctorj]

CanOz, good work at ignoring the childish bickering and continuing with the thread.

I will convert to Metastock over the weekend and have a play myself.

In the mean time, for my own curiousity, are you able to run it on some charts to get Amibroker to identify the situation historically, so we can begin to identify characteristics of the common moves resulting from this set up and post them here with a view to begin analysing how best to manage the exits or alternatively looking if we need to expand the entry conditions.

Thanks for your great work.

 

[CanOz]

CanOz, good work at ignoring the childish bickering and continuing with the thread.

I will convert to Metastock over the weekend and have a play myself.

In the mean time, for my own curiousity, are you able to run it on some charts to get Amibroker to identify the situation historically, so we can begin to identify characteristics of the common moves resulting from this set up and post them here with a view to begin analysing how best to manage the exits or alternatively looking if we need to expand the entry conditions.

Thanks for your great work.

Hi Doc. Yes i can run LESM's code on historical ASX, all data, but not until i get back home or get time to finish my laptop setup here at work. I'll try to get my laptop setup properly today with P.Data and Amibroker.

Thanks LesM for the code!

I only did the exploration on the EOD from the 31st as i was getting ready for work this morning, should have more time at home tomorrow morning too. It pulled up 1/2 dozen or so and i just picked the pattern that was the most defined. Obviously i didn't filter out the smaller stocks either.

So am i right to say that we're looking for little micro triangles, showing this decrease in volitility?

Cheers,

 

[doctorj]

So am i right to say that we're looking for little micro triangles, showing this decrease in volitility?

Lets start with this and see how we go. Once we study historical entries, we should get a feeling for if its got the potential to be valid.

With these small triangles, how do we define whether they're a long or short entry? Should we look to use an MA or 2 to determine if they're in an up or down trend and take that side of the trade accordingly? What other tools do we have to define a trend?

 

[tech/a]

doc.
add a condition to trigger a buy or sell.

 

[CanOz]

Lets start with this and see how we go. Once we study historical entries, we should get a feeling for if its got the potential to be valid.

With these small triangles, how do we define whether they're a long or short entry? Should we look to use an MA or 2 to determine if they're in an up or down trend and take that side of the trade accordingly? What other tools do we have to define a trend?

Yeah maybe an EMA to start with.....or highest highs, lowest lows etc.

Lets get some input on this?

I don't think we need the buy or sell signal yet do we? As long as we can run it in exploration in Amibroker....do you need the signal in MS?

Cheers,

 

[lesm]

CanOz, good work at ignoring the childish bickering and continuing with the thread.

I will convert to Metastock over the weekend and have a play myself.

In the mean time, for my own curiousity, are you able to run it on some charts to get Amibroker to identify the situation historically, so we can begin to identify characteristics of the common moves resulting from this set up and post them here with a view to begin analysing how best to manage the exits or alternatively looking if we need to expand the entry conditions.

Thanks for your great work.

Can,

As doc mentioned, this gives a starting to point to investigate and identify the characteristics in more detail. We can then work out an approach and way forward from there and see what eventuates.

Cheers.

 

[R0n1n]

While googling for system development I found this. Looks good. May be we can use it as a framework/guidelines to work with.

 

[Nick Radge]

The decrease in volatility is an entry condition. The market will dictate whether the entry will be long or short so we just need to be ready to travel with it.

The initial question is whether one wishes to trade in the direction of the current trend or whether it doesn't matter either way. If the system is short term, i.e. out to the 3-days I'd suggest it will not matter. However, if you are looking for some longer moves then perhaps in the direction of the prevailing trend. If so, one needs to define the current trend.

Next, we need an entry trigger. Wayne has mentioned an ATR as an exit. Well, and ATR can also be used as an entry as well.

So, once the entry condition is met we then need to use a reference point to measure off the entry level. We can use the highest high or the lowest low of the contraction period or we can use the high or low of the most recent bar.

As an example:

If entry condition met then look for entry trigger;
If high is exceeded by n ATR then enter long;
If low is exceeded by n ATR then enter short;

In other words we want to enter in the direction the market moves after the volatility contraction. We simply want to "hop on the train as it leaves the station"

 

[tcoates]

Just to add something to the mix for consideration (following from what Nick was saying)...

http://www.traderslog.com/Volatility-Breakout-Systems.htm

Tim

 

[Captain_Chaza]

.."Sailing by the Book "

There are many sailing techniques men have used in their quest for the Holy Grail. The most difficult task is deciding which ships to sail and then know when to rest. Knowing when to rest is the hardest.

Here is simple approach you might try if you are sailing the ASX

1. Determine the overall global condition by surveying a folio of charts in the Global Exchange
(Wall St)

2. Determine the classes of ships best suited to these conditions by surveying a folio of charts in Sailing against the All Ords

3. Design a craft or a small fleet best suited to your own needs and comfort.

4. Then set sail using logical seamanship disciplines that minimizes risks and maximizes opportunities.

Or

1. You pick the sails using either a company or industry you are familiar with, or the recommendation of a trusted analyst (either fundamental or technical).
Remembering ALWAYS "It is the Sailor and not the Sail"

2. Determine the seaworthiness of the craft by comparing her chart with others.

3. Then set sail using logical seamanship disciplines that minimizes risks and maximizes opportunities.

Whichever approach you take, most importantly the fine balance between safety and adventure can only be learnt with experience and lots of practice drills and ocean trials.

Much of your success with these charts will come from the attention given to ocean trials.
Hopefully “Ocean Trials” and Practice Drills will be as much fun as it will be constructive in achieving the high degree of seamanship necessary nowadays when setting sail on any member of the Global Exchange

With today’s sophisticated, computer models and extensive tunnel testing approach to ship building and design there is still no better test of a ship’s performance and seaworthiness unless it is compared to a few others.

All in the same seas, all under the same weather conditions, all at the same time

That is, A RACE !

By plotting past voyages on these maps, experienced skippers will now be able to survey their own sailing techniques and very quickly recognize if there is any scope for improvement.

Surprisingly, you don’t need to know how to sail to know if it is a good day to go sailing.

If you sit by the window in a high rise office building in view of Sydney Harbour or Port Phillip Bay on some days you will see a few yachtsman setting their sails and then followed by many more.

On other days there will be no one in sight.

Most days however some will be heading North bound skippered by the bullish conditions and others favour the Southerly course. Some will be set by a calm.

Whichever direction you prefer to set sail it must be remembered that each craft will behave in her own way under the same circumstances and it may be quite different than any other vessel.

What must be forgotten are the old sailors tales "She'll be right in the long term"

As you zoom in with binoculars on some days you will find an only few classes of yacht are out there amongst the elements.

When you see them with a full spread of sail presented to the wind you can easily appreciate my addiction to this great sport.

Bon Voyage
Captain Chaza

 

[CanOz]

The decrease in volatility is an entry condition. The market will dictate whether the entry will be long or short so we just need to be ready to travel with it.

The initial question is whether one wishes to trade in the direction of the current trend or whether it doesn't matter either way. If the system is short term, i.e. out to the 3-days I'd suggest it will not matter. However, if you are looking for some longer moves then perhaps in the direction of the prevailing trend. If so, one needs to define the current trend.

Next, we need an entry trigger. Wayne has mentioned an ATR as an exit. Well, and ATR can also be used as an entry as well.

So, once the entry condition is met we then need to use a reference point to measure off the entry level. We can use the highest high or the lowest low of the contraction period or we can use the high or low of the most recent bar.

As an example:

If entry condition met then look for entry trigger;
If high is exceeded by n ATR then enter long;
If low is exceeded by n ATR then enter short;

In other words we want to enter in the direction the market moves after the volatility contraction. We simply want to "hop on the train as it leaves the station"

Thank Nick, we've got some things to work on for a day or so anyway...i'll try to get an example coded up tomorrow....Tech you wanted the signals so we can start testing?

LESM and Doc you'll probably beat me to it in the coding.....

Should we code up an exit now as well just to test?

Cheers,

 

[Captain_Chaza]

The decrease in volatility is an entry condition. The market will dictate whether the entry will be long or short so we just need to be ready to travel with it.

The initial question is whether one wishes to trade in the direction of the current trend or whether it doesn't matter either way. If the system is short term, i.e. out to the 3-days I'd suggest it will not matter. However, if you are looking for some longer moves then perhaps in the direction of the prevailing trend. If so, one needs to define the current trend.

Next, we need an entry trigger. Wayne has mentioned an ATR as an exit. Well, and ATR can also be used as an entry as well.

So, once the entry condition is met we then need to use a reference point to measure off the entry level. We can use the highest high or the lowest low of the contraction period or we can use the high or low of the most recent bar.

As an example:

If entry condition met then look for entry trigger;
If high is exceeded by n ATR then enter long;
If low is exceeded by n ATR then enter short;

In other words we want to enter in the direction the market moves after the volatility contraction. We simply want to "hop on the train as it leaves the station"

I think you are on the right tack here, Capt Radge

It is always diificult to judge the difference between a ship Beset by a Calm and ONE waiting patiently to break out of the Doldrums in any direction

Salute and Gods' Speed

 

[tech/a]

Ill knock up some code if I get a chance.

 

[Captain_Chaza]

Ahoy Officer Radge
I ask you
Technically speaking?

Once you get to the stage of a Calm
The Volitility/Lack of volitilty you crave for is LOST ?
Where then ???

Once you get tothe Doldrums
The Volitility/Lack of volitilty you crave for is LOST AGAIN?

Where then???? Do we set our next course????

Salute and Gods Speed

 

[CanOz]

Ahoy Officer Radge
I ask you
Technically speaking?

Once you get to the stage of a Calm
The Volitility/Lack of volitilty you crave for is LOST ?
Where then ???

Once you get tothe Doldrums
The Volitility/Lack of volitilty you crave for is LOST AGAIN?

Where then???? Do we set our next course????

Salute and Gods Speed

For sure we're going to this happen, we just need to decide what to do if its an issue, it should show up in the testing.

Cheers,

 

[theasxgorilla]

An update on IDO...

Somethings to ponder (only as an example):

* The price exceeded the high of the first bar of the pattern, $1.25, by 1 cent. This is one possible entry trigger (TRIGGER 1), as per Nick's mail above.

* The price also (of course) exceeded the high of the last bar in the pattern, $1.23, another possible entry trigger (TRIGGER 2).

* If you assume the profit target for such a triangle to be 1x the range of the 1st bar as extended from the apex, then the profit target here is $1.30 (assuming an apex of $1.22).

* If you have a tight stop 1c below the last bar of the pattern ie. technically outside the pattern, at $1.20, then a win (reaching the profit target) using TRIGGER 1 would be 0.5R, and using TRIGGER 2 would be 1.2R.

Of course this is one instance, not the universe. In anycase, for mine, the more aggressive entry (TRIGGER 2) is probably better. Swing for the fences!

 

[theasxgorilla]

I think you are on the right tack here, Capt Radge

It is always diificult to judge the difference between a ship Beset by a Calm and ONE waiting patiently to break out of the Doldrums in any direction

Salute and Gods' Speed

This is true, and in back testing the outcome of statistics such as win/loss ratios in combination with an average winning trade to average losing trade will show (historically) if you're system 'judges' well enough to be profitable.