Autotrading, is it really...

Questions about MultiCharts and user contributed studies.
User avatar
Januson
Posts: 119
Joined: 18 Apr 2007
Location: Denmark

Autotrading, is it really...

Postby Januson » 05 Aug 2008

Hello all dear MC-users 8)

I'm sitting here again, first I did some searching for text placement inside a chart and a lot of threads matched my search criteria - but couldn't quite find what I was looking for. Then I thought to myself.. huh, why not go through the last couple of release notes for MC.

So I did...
I realized there has been and still are an enormous effort to create autotrading features.
I also know there has been polls for feature request etc., but I'm still amazed that so much development time has been used trying to fulfill the users requests - Which is very nice.. don't get me wrong here :D

However, I'm very curious to know how many users actually use autotrading in MC and if so.. is there any traders that earn their money using it?

I know TSSUPPORT is going the demochratic way using polls and so forth - which is very nice (once again)
BUT I must admit I'm having a darn hard time trying to understand why small feature requests is being totally ignored. IMHO those autotrading features and optimization routines (I've never really understood optimization, another time though ;) ) is much more difficult to develop than the ones needed for us discretionary traders :lol:

I could mention:

1 ) Practice trading with a simulated datafeed allready present i QuoteManagers Cache. NT could be a nice inspiration.

2) Email notifications for signals, should be fairly easy and nice for short/ long term traders.

3) Text floating inside chart windows, either the superior way as MT4 take a look at the attached screenshots or it could be the simple way just by using a floating textbox.

4) Save all...

5) Rearrange tabs, 1 lines or 2 lines of tabs instead of using arrows


I have omitted strategy building, expert commentary etc. I believe the above points should cover the most discretionary-traders.

I hope I haven't offended anybody with this, but I think the polls should be constructed in such a way that when feature requests for automated trading has been at the top and has been filled by devteam, they should be downgraded for a while so that new (SMALL) features could pop up also :roll:

Cheers and nice trading 8)
Attachments
sqo3cy.gif
(17.54 KiB) Downloaded 2106 times
8c0zev6.gif
(14.42 KiB) Downloaded 2106 times

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 05 Aug 2008

Januson,

Autotrading is important because MultiCharts is boldly offering all and sundry to:

Unleash the Power of Algorithmic & Automated Trading THROUGH
MultiCharts WHICH is a professional charting and automated trading platform, designed for futures, stock, and forex traders. Develop, back-test, and optimise your trading ideas.

The words in CAPITAL are mine, others are by MultiCharts.

For 23 years I have made my own trading decisions based on technical analysis; I am still a discretionary trader. And I do not foresee stopping discretionary trading altogether. But I also wanted to explore automated trading, even way back in 1980's. The technology was not within my reach.
It was after I purchased MultiCharts that I became aware I could actually translate the ideas I had into a software and run statical analysis, determine net advantage and thus determine optimum position sizing and control risk.

However, it has been mostly frustration because MultiCharts is taking its sweet time breaking the exclusivity with IB.

Yes, I know there is a free SDK and a 5K SDK facilitating writing the writing autotrading links for brokers other than IB, but in my humble opinion such is a last resort, and not recommended, solution.

There are many other matters which has inhibited development of trading techniques which I shall address tomorrow under the TRADE SIMULATION topic.

I know you are a strong MultiCharts supporter. So am I. I have rooted for MultiCharts, and defended it against unjustified attacks, on this forum and on kreslik.com. These are the only two forum I frequent.

Khalid

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Re: Autotrading, is it really...

Postby Marina Pashkova » 07 Aug 2008



BUT I must admit I'm having a darn hard time trying to understand why small feature requests is being totally ignored. IMHO those autotrading features and optimization routines (I've never really understood optimization, another time though ;) ) is much more difficult to develop than the ones needed for us discretionary traders :lol:
Hi Januson,

Small features are not getting ignored. It's all a matter of priorities. 90% of our development time is dedicated to the new large features that make MultiCharts more advanced and suitable for solving a wide range of tasks. 10% of our development time is dedicated to the small features that do not change the product in any dramatic way. They do improve the platform's usability, but they are not critical. Hence the time allocation proportion.

1 ) Practice trading with a simulated datafeed allready present i QuoteManagers Cache. NT could be a nice inspiration.
Already planned
2) Email notifications for signals, should be fairly easy and nice for short/ long term traders.
Planned to be implemented in the near future.
3) Text floating inside chart windows, either the superior way as MT4 take a look at the attached screenshots or it could be the simple way just by using a floating textbox.
If I understand you right, what you're talking about can be done in MC.
4) Save all...
Could you elaborate on that please?
5) Rearrange tabs, 1 lines or 2 lines of tabs instead of using arrows
You can choose between showing tabs either on top or on bottom as well as rename them.

I have omitted strategy building ...
Could you elaborate on that please?

Regards.

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 07 Aug 2008

However, it has been mostly frustration because MultiCharts is taking its sweet time breaking the exclusivity with IB.
Hi Khalid,

We have finished working on FXCM as a supported broker. Our engineers have almost completed developing Zen Fire and are in the process of developing Pats. Trading Technologies is next in the line.

Regards.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Re: Autotrading, is it really...

Postby Tresor » 07 Aug 2008



I have omitted strategy building ...
Could you elaborate on that please?

Regards.
Hi Marina, please see this link: http://www.tradecision.com/product/trad ... uilder.htm

Regards

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Re: Autotrading, is it really...

Postby TJ » 07 Aug 2008

3) Text floating inside chart windows, either the superior way as MT4 take a look at the attached screenshots or it could be the simple way just by using a floating textbox.
If I understand you right, what you're talking about can be done in MC.

Regards.
it can be done... it is not easy. it is not a "fill-in-the-blanks" thing. it will take some programming to do.
there must be a pre-canned version out there somewhere.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 07 Aug 2008

Tresor,
I wish strategy development was as easy as pointing and clicking a mouse; this is not strategy building but repackaging of indicators.

At times I am harsh with TS SUPPORT, but during the more than two years I have been using MultiCharts I am very happy to say I have not come across one instance of such irresponsible behaviour by TS SUPPORT.

That is why, despite my many frustrations, I intend to remain an active MultiCharts user.

Khalid

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 07 Aug 2008

Tresor,
I wish strategy development was as easy as pointing and clicking a mouse; this is not strategy building but repackaging of indicators.

At times I am harsh with TS SUPPORT, but during the more than two years I have been using MultiCharts I am very happy to say I have not come across one instance of such irresponsible behaviour by TS SUPPORT.

That is why, despite my many frustrations, I intend to remain an active MultiCharts user.

Khalid
Yes, but to repack indicators in MC you need to be a programmer. A repacking of indicators in Tradecision can be made by 5 year old child.


Regards

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Postby TJ » 07 Aug 2008

Yes, but to repack indicators in MC you need to be a programmer. A repacking of indicators in Tradecision can be made by 5 year old child.

Regards
I almost want to say: "Don't be naive...". But it is not my place to judge, because I am naive too.

if auto-trading strategy development is SO easy... there will be ONLY one product on the market -- the Tradecision.

Correction, actually there won't be a product called Tradecision. Because whoever got the first copy would have bought out the company and prevent anybody to have a copy.

The reality is...
The market is dynamic. (i.e. it is ever changing)
The market is sophisticated. (i.e. it has a mind of its own)
The market is complicated. (i.e. it has multiple personalities)

To reduced the above three characteristics of the market into a simplistic auto-trading algorithm has been a dream of many.

... and for some, it is a life-wasting dream.

One thing I can be certain, and many will concur (this might save you hours of backtesting and development work): a profitable auto-trading strategy is more than a combination of MACD, Stoch, CCI, etc. If the Holy Grail lies within these indicators, we would all be rich years ago.

Another thing I can be certain of... there will always products on the market to feed this dream... the dream of creating the Holy Grail.

Have a good journey.

TJ


p.s. this is not to say, profitable auto-trading does not exist. What I am alluding to is, I doubt anybody can come up with a long term profitable auto-trading strategy by simply combining pre-canned indicators.

p.p.s. if you are a logical person (which I think you are. Otherwise you would not even dream of auto-trade algorithm), if you can articulate your logic in your own language, if you can write down your logic on paper... YOU CAN PROGRAM in EasyLanguage.
If anybody who does not agree with my statement, please feel free to scold me for making such a bold and irresponsible statement.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

WHY REPACK INDICATORS?

Postby khalaad » 08 Aug 2008

Tresor,

Why would one want to repack indicators?

Because it is the easy way out?

If you are lucky, such a system may land you a few windfall profits.

Eventually, not matter how you optimise the thing -- and sooner rather than later -- THE SYSTEM WILL WIPE OUT YOUR ACCOUNT.

TJ,

You have said it very well. God bless you.

Khalid

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 08 Aug 2008

Hi Khalid and TJ,

Khalid, I rate high your trading comments on kreslik.com. You are a trading guru for me. When you look at the Strategy Builder you say ‘’The system will wipe out your account’’. TRUE.

TJ, I read many post of yours on programming and MC. You are a programming guru for me. When you look at the Strategy Builder you say ‘’You can program (it yourself in) EasyLanguage’’. TRUE

My background is in marketing. When I look at the Strategy Builder I see many commercial opportunities for TSSupport.

We look at the issue from different viewpoints, therefore no compromise can be achieved in this respect:

Let me introduce 2 conflicting marketing terms to you:

1. ‘’to enhance user experience’’. If a producer fails to keep up what the customer wants, the producer might lose the customer (in many cases such a producer will go out of the business). If you have always driven Mazda car and you liked this make, but for some reason decided to switch from a manual gear box to an automatic one and Mazda does not offer automatic gear boxes in their cars, you will buy a Toyota with automatic gear box = Mazda will lose the customer.

If you are a successful discretionary trader and want to try algorithmic trading and you code your strategy but – due to limited programming skills – you will fail to code, e.g. ATR multiplier for a trailing stop, wouldn’t it be nice if you simply clicked your money management wizard built-in the Strategy Builder and let the machine do all the coding for you? 50% of users will do further education on programming and will program the trailing stop by themselves, 50% will go for software that offers them satisfactory user experience and will utilize the time they saved for trading rather than for programming. See my point :-) ?


2. ‘’Disruptive innovation vs. sustaining innovation’’ this term will be a bit complex than the above. Just google this term or find in Wikipedia to get a better understanding.

Let’s assume we are in the back 20 or 30 years and we are producers of photographic films for amateur photo cameras. We are one of 300 producers of such materials in the world. We are doing fine. Clients demand higher resolution? We deliver higher resolution films. They demand colour films? We provide colour films and are happy, so are our clients. We are constantly improving our product (we do sustaining innovation) and think we have done our job well. Do we run our business properly? Most of us will say ‘’YES’’. Of 300 hundreds producers of photographic films there remained only 3 – 5 now. We got killed by a disruptive innovation, i.e. by a digital photography.

Now guys, let’s pretend we are a charting software producers (one of 30, 50, 100 – I have no idea of how many developers of charting and analytical software there are n the world). Clients ask us for an optimization tool? We provide them with it! Clients ask for a play-back utility? We provide them with it. Clients ask for many more additional features? We provide them with it! Do we run our business well? Most of us will say ‘’YES’’.

Clients ask for a STRATEGY BUILDER – should we provide this feature to them? Try to think how the STRATEGY BUILDER could be used as a disruptive innovation (a breakthrough) rather than a sustaining innovation (additional feature) and you will know the answer :-)

Bear in mind two issues while answering this question:

(i) it often happens that a company that devices a disruptive technology goes out of business simply because they committed one silly mistake (e.g. Pandora.com devised a customized radio business and served / serves perfect music, but last.fm based their customized music serving engine on listeners’ generated preferences and therefore succeeded – the business was sold by 3 students for GBP 140). Isn't disruptive innovation fine if done properely? If Strategy Builder concept devised by Tradecision is to be a disruptive technology then it needs to be reshaped into what I call ‘’a reversed strategy builder’’ – sorry for this enigmatic phrase.

(ii) the following paradigm that always works in business (no matter of line of your business): when a disruptive innovation comes to life, more than 90% of existing businesses / technologies will go out of the market.
Once again, please try to think of how a different (than Tradecision’s) approach could be used with Strategy Builder and you will know how to kill MC’s competitors.

Khalid and TJ, I do not want to argue with you about trading – I have little to say about this, because I am a newbie in this area. What I merely did was to show Marina a link to a strategy builder with which she was not familiar with.

I am, like you, a supporter of MC. But from my non-trading point of view a chasing of competitors with adding additional features is okay but only for short-term. In a longer term this chase will be fruitless when a disruptive innovation comes.

Is adding the modified Strategy Builder just a chase for additional features? I personally believe that Strategy Builder, after reshaping, can be a long-term asset for MC / can be a disruptive innovation.

Regards

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 08 Aug 2008

Have a nice weekend guys :D

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Postby TJ » 08 Aug 2008

Tesor:

Thanks for your thoughts.
We are not arguing... sharing ideas will only help us understand each others needs, and at the end, help MC to produce a better product for all.

;-)

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 08 Aug 2008

Tresor,
My background is in marketing. When I look at the Strategy Builder I see many commercial opportunities for TSSupport.
These 'commercial opportunities' will be only to short term advantage and do great damage to TS Support integrity and its aspiration of becoming a serious player in it chosen area of business.
If you have always driven Mazda car and you liked this make, but for some reason decided to switch from a manual gear box to an automatic one and Mazda does not offer automatic gear boxes in their cars, you will buy a Toyota with automatic gear box = Mazda will lose the customer.
People shall stop buying any car which is accident prone. Similarly if TS Support is greedy enough to ape Tradecision it will assist people in loosing their money and quite likely eventually go out of business.
If you are a successful discretionary trader and want to try algorithmic trading and you code your strategy but – due to limited programming skills – you will fail to code, e.g. ATR multiplier for a trailing stop, wouldn’t it be nice if you simply clicked your money management wizard built-in the Strategy Builder and let the machine do all the coding for you? 50% of users will do further education on programming and will program the trailing stop by themselves, 50% will go for software that offers them satisfactory user experience and will utilize the time they saved for trading rather than for programming.

The key here is 'satisfactory user experience'. You are assuming what Tradecision is offering is good enough to offer 'satisfactory user experience'. Or that TS Support can come up with something offering 'satisfactory user experience'. It is self evident you do not know this but this gimmick by Tradecision is a rehash of similar earlier gimmicks by other irresponsible/dishonest purveyors of instant riches I seen since 1985 none of whom have survived and all of whom caused lot of financial hardship to those gullible enough to believe they will save time and use it for trading.

While I accept a technological breakthrough is always possible and a product offering 'satisfactory user experience' is quite conceivable such will probably be priced in tens of thousands of dollars.
Khalid and TJ, I do not want to argue with you about trading – I have little to say about this, because I am a newbie in this area. What I merely did was to show Marina a link to a strategy builder with which she was not familiar with.

Your background is marketing yet you are proposing this become-the-industry-leader grand scheme without proper market research; what seems a novel idea and technological marvel to you has been around for at least twenty years in various sahpes and forms, and I am quite certain Marina and TS Support are aware of what Tradecision is up to and what Tradecision's role models were up to and where they ended up.

Khalid

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

OOPS!

Postby khalaad » 08 Aug 2008

Tresor,

Oops, I messed up the post. A perfect illustration in how one can quite easily go wrong with technology one is not fully familiar with!!

My background is in marketing. When I look at the Strategy Builder I see many commercial opportunities for TSSupport.


These 'commercial opportunities' will be only to short term advantage and do great damage to TS Support integrity and its aspiration of becoming a serious player in it chosen area of business.

If you have always driven Mazda car and you liked this make, but for some reason decided to switch from a manual gear box to an automatic one and Mazda does not offer automatic gear boxes in their cars, you will buy a Toyota with automatic gear box = Mazda will lose the customer.


People shall stop buying any car which is accident prone. Similarly if TS Support is greedy enough to ape Tradecision it will assist people in loosing their money and quite likely eventually go out of business.

If you are a successful discretionary trader and want to try algorithmic trading and you code your strategy but – due to limited programming skills – you will fail to code, e.g. ATR multiplier for a trailing stop, wouldn’t it be nice if you simply clicked your money management wizard built-in the Strategy Builder and let the machine do all the coding for you? 50% of users will do further education on programming and will program the trailing stop by themselves, 50% will go for software that offers them satisfactory user experience and will utilize the time they saved for trading rather than for programming.


The key here is 'satisfactory user experience'. You are assuming what Tradecision is offering is good enough to offer 'satisfactory user experience'. Or that TS Support can come up with something offering 'satisfactory user experience'. It is self evident you do not know this but this gimmick by Tradecision is a rehash of similar earlier gimmicks by other irresponsible/dishonest purveyors of instant riches I have seen since 1985 none of whom have survived and all of whom caused lot of financial hardship to those gullible enough to believe they will save time and use the saved for trading.

While I accept a technological breakthrough is always possible and a product offering 'satisfactory user experience' is quite conceivable such will probably be priced in tens of thousands of dollars.

Khalid and TJ, I do not want to argue with you about trading – I have little to say about this, because I am a newbie in this area. What I merely did was to show Marina a link to a strategy builder with which she was not familiar with.


Your background is marketing yet you are proposing this become-the-industry-leader grand scheme without proper market research and without a clear understanding that a software version of Alladin's lamp is never going to help any trader BECAUSE THERE ARE NO SHORTCUTS TO ACHEIVING FINANCIAL FREEDOM VIA TRADING. What seems a novel idea and technological marvel to you has been around for at least twenty years in various sahpes and forms, and I am quite certain Marina and TS Support are aware of what Tradecision is up to and what Tradecision's role models were up to and where they ended up.

Khalid

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Postby TJ » 08 Aug 2008

I hesitate to add more to this thread, because very often good natured discussion appears confrontational in a written forum.


I was in marketing for a MNC.
One of my products is recognized as a World leader, and is widely implemented in Exchanges around the World.

We are in an imperfect world...
There are products that are useful...
and then there are products that sell...

We can go into detail later... Better still, when we meet for a beer, we can talk about the zen of system trading.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 08 Aug 2008

Khalid,

Please read my post again.

1. I am NOT anywhere assuming Tradecision's Strategy Builder is offering satisfactory user experience.In fact what I claim is that the concept of Strategy Builder was screwed up by them and would need to be be shaped other way.

2. I introduced the term ''enhancing user experience'' which is adding additional features, which is good for customers short term, but may lead to a developer's commercial disaster long term (90% companies satisfying their customers by doing this go out of business when disruptive innovation comes to life). Didn't I write that?

3. I introduced the term ''disruptive innovation'' which - if executed properly - leads to a long term success.

4. How come you thought I would favour the solution mentioned in point 2?

5. You totally missed my point.

Regards

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 08 Aug 2008

Yes.

I have probably said more than how much, and what, is appropriate on a public forum.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 08 Aug 2008

We are in an imperfect world...
There are products that are useful...
and then there are products that sell...
TJ, FULLY AGREE

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 08 Aug 2008

Yes.

I have probably said more than how much, and what, is appropriate on a public forum.
Let's chill out during weeknd :mrgreen:

Nick
Posts: 496
Joined: 04 Aug 2006
Has thanked: 4 times
Been thanked: 24 times

Postby Nick » 13 Aug 2008

My two cents....MC is unsuitable for auto trading for a variety or reasons. Just off the top of my head...tick filter ...holiday management....General robustness. If you want to run auto - run it on something like strategy runner on your brokers server in a properly managed data centre.

Personally I'd like to see more effort on 'core' features like data management or genuinely useful stuff like tick precise technology.

I have nothing against algorithmic trading but MC is just not the package you would pick for that task. (imo of course).

I'd be interested is anyone running 24/7 completely un-attended (i.e. in a managed facility) with MC?

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 14 Aug 2008

My two cents....MC is unsuitable for auto trading for a variety or reasons. Just off the top of my head...tick filter ...holiday management....General robustness.
Hi Nick,

Could you elaborate on the above please?
Personally I'd like to see more effort on 'core' features like data management or genuinely useful stuff like tick precise technology.
Could you please elaborate on what changes you would like to see in data management or what you mean by the 'tick precise technology'?

Thank you.

Nick
Posts: 496
Joined: 04 Aug 2006
Has thanked: 4 times
Been thanked: 24 times

Postby Nick » 14 Aug 2008

Its just I wouldn't leave MC (or any 'desktop' charting package) running while I went on holiday. It might be a great tool for prototyping but to run live strategies on a live account I would want a whole lot more. I would want a high level of redundancy (i.e. multiple instances of the system running in case one fails) and to be able to recover gracefully from errors (wherever they occur) i.e. to work out what the current market position should be and adjust the portfolio accordingly. I would personally either pay a broker to manage a system or use technology that has a demonstrable (and guaranteed) high availability.

Anyway I certainly would wish anyone well that is going down that path and would love to hear how it pans out :)

'Tickprecise' is a name coined by Neoticker. It is a fantastic feature that allows historical data to be loaded in an identical fashion to live data. So een if you are loading say 5 minute bars it builds them tick by tick. It will also do this with multiple data streams. A simple application for this would be MD studies that work on historic data (volume@bid/volume@ask). It also has thepotential of greatly increasing te accuracy of backtesting.

Incidentally I believe that NT are promising something similar for their next release version 7.

An example application - construct a study that takes order book changes and actual order flow (trades) as data streams - compare pulled orders against filled orders. That could be done without 'tick precise' by working on it 'live'. But to do any sort of meaningful analysis and refinement you would need to have it work on historical data. This would mean that ticks from the various data streams would need to be fed into MC in the correct order with the correct time stamp.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 15 Aug 2008

Nick,

What we ordinary people are talking here is computer-assisted trading rather than FULLY automated trading.

As long as I have properly enabled my computer with MultiCharts to follow and implement what I consider profitable trading method(s) I would rather trust it than myself because:

I need not concentrate on every trading decision and would remain free to do other things;
the computer does not suffer from lapses of concentration;
the computer does not get that disorientating phone call;
the computer does not have moods;
the computer does not tire;
the computer, when sick, is replaceable within minutes with a healthy backup;
etc, etc.

Anyone who has just a little awareness in these matter should know UN-ATTENTED and PROFITABLE FULLY automated trading is not possible with MultiCharts. Or for that matter with Neoticker, NT, etc.

For SUSTAINABLE, PROFITABLE, UN-ATTENTED, FULLY automated trading one would be advised to have a bespoke application which with as few black boxes between the application and the market. None of the worthy neos and ninjas, however sophisticated and impressive otherwise, work the way I would want them to work and are not yet capable of handling the nuances in trading to be left completely unattended.

Khalid

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 15 Aug 2008

Nick,
Its just I wouldn't leave MC (or any 'desktop' charting package) running while I went on holiday.
True.

Personally, I would hang around under the same roof!

Khalid

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Postby TJ » 15 Aug 2008

What we ordinary people are talking here is computer-assisted trading rather than FULLY automated trading.
Well said.


Furthermore,
Unattended 24/7 Autotrading was NEVER intended. (rhythms nicely.)

Not by Goldman Sachs, or any trading houses large, medium or small.


First, most of us DO NOT have the technology to enable unattended autotrading. MultiCharts is only one small part of the totally equation -- we will need end-to-end redundance of everything for a "No-Single-Point-of-Failure" setup.
[edit] i.e 2 computers, 2 installation of MC, failover software, dual network card, dual router, dual internet connection, etc., also important -- your MC code is failover aware and failure recovery able.

Secondly, we have the wrong broker to implement Unattended 24/7 Autotrading. Because IB reboots its server at midnight.

Thirdly, all the trading houses that employ autotrade, have teams of specialists scrutinizing the system every second of its 24/7 operation -- from the computer administrators who make sure the computer works, to the network administrators who make sure the internet connects, to the analyst who make sure there is no unexpected news, to the risk management officer who make sure the exposure is not out of line...

They were never unattended !!!

This is the reality.
Last edited by TJ on 15 Aug 2008, edited 6 times in total.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 15 Aug 2008

TJ,

It is always a pleasure to read someone who knows what he/she is talking about and has also done the homework.

So rare.

Khalid

Nick
Posts: 496
Joined: 04 Aug 2006
Has thanked: 4 times
Been thanked: 24 times

Postby Nick » 15 Aug 2008

This is what I figured, however I do wonder what the appeal of 'assisted' trading is. If you have a profitable mechanical strategy hand it to a (trusted) broker to trade - or run it on something like strategy runner where they have the redundant infrastructure in place already.

Personally I would like to see 'primary' features get the most attention from MC. That's with no disrespect to you auto guys but if you can't go off sailing or to play golf what's the point? Most broker platforms will ratchet stops up for you on open positions.

I do know a couple of retail traders that have gone 'algo' but they have rolled there own (using open source tools, code, and applications). They rent server space in a managed data centre to run things. Pretty cheap nowadays.

Anyway I certainly wish you guys all the best in your endeavours and would certainly like to hear about your progress (if you feel like posting the odd update). :D

Cheers,
Nick.

Nick
Posts: 496
Joined: 04 Aug 2006
Has thanked: 4 times
Been thanked: 24 times

Postby Nick » 15 Aug 2008

Just to reminisce a little. I used to be involved with a company that put in high availability systems some 30 years ago now. They where based on DEC's PDP 11 and where fully replicated (locally and at separate locations). These where turnkey systems (no operating system to screw up). Of course the complexity of the platforms we run on now (windows in particular) do not lend themselves to fault tolerance.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 17 Aug 2008

Nick,

I think the purpose of this forum is free exchange of ideas, information, etc.

It is NOT to persuade others to agree with a certain point of view.

Moreover, I believe it rather impertinent asking strangers about their trading performance(s)?

So let's agree to disagree and leave it at that.

Khalid


Return to “MultiCharts”