How the Problems of Photography relate to Forex Trading and what to do about it

[marc walton]

Xmas 2011 Competition Entry by Julian Jackson. Julian is always an enthusiastic & willing participant in the forum & in competitions. Here is his offering for this years & I am hoping to persuade him to take a more active role in the forum & carry this interesting thesis forwards in to the new Year and beyond.
A few days ago I was photographing tarmac. Yes, tarmac. It was unexpectedly more difficult than I thought. Before you start wondering about my sanity, not to mention my photographic sensibilities, I should tell you that it is for a childrens’ playground I have raised funds for. The image was to send to council officials to show them the area I propose to have the project built on.

I was using my one-year old digital camera, a Canon EOS 60D, which is an extremely capable semi-professional digital SLR. It was a significant investment for me, but a necessary one, as inferior tools produce second-rate results.

When I tried to take the photo, theautofocus fail indicator blinked in the viewfinder. I was a bit surprised, and checked the settings on the camera, and tried again. No good. In broad daylight the autofocus was unable to establish the distance and was “hunting” – moving around looking for the focus.

I’ve probably taken 1000 pictures now with this camera and the autofocus is extremely good, except in low light or featuring reflective surfaces like mirrors or water – known problem areas.

After a moments thought I knew what had happened: the autofocus works by an infrared beam reflected back from the subject. In this case the black tarmac must have absorbed most of the beam and not enough made it back to the camera for it to work out the distance to focus on. I glanced around for something else to focus on and took the picture.

Over the many images I have taken with this camera, in differing light conditions, of fast moving subjects and closeups, landscapes and portraits, the camera’s electronic systems have been incredibly accurate. Over the last 30 years I have owned quite a few cameras and while lenses, batteries, and mechanical components have improved slightly, the computerisation and electronic capabilities of cameras have developed astonishingly, as well as moving from film to digital image capture.

Without the accuracy of the indicators the camera provides, I wouldn’t be able to take good – that is within acceptable technical parameters – images. Leaving aesthetics aside, if you can’t capture a properly exposed and focussed image, there’s no point bothering. A few years ago I bought a cheap camera for “Happy Snaps” and it had such bad “shutter lag” – that is the time between when you press the shutter button and when the image is captured, that I was glad when it was stolen! As in forex trading, lagging indicators are near useless.

What would happen if the focus of my camera was too short one day, then over the next day, or the exposure wobbled around by a few f-stops each time, and sometimes, depending on the light, was completely wrong? If shutter speed and lens aperture no longer seemed related? I’d be reduced to guesswork each time I tried to shoot an image.

Isn’t this the situation we face in forex trading? The lack of accurate indicators reduces us to what in my view is often no more than educated guesswork, which in turn means missed opportunities, and flawed entries and exits?

Some of the common indicators we use are 30 or more years old. Can they give us the best picture possible of the market state? Which of you uses a 30 year old computer or software? Even lurching back to my camera metaphor, while there are some excellent 30+ year old cameras, I wouldn’t use them on a mission-critical shoot.

Some of you may be thinking that I am comparing apples to oranges. Yes, I am using a metaphor and it shouldn’t be taken too far, but my hypothesis is that we in the FMP community are not making best use of the huge leaps in computing and mathematical applications that have taken place over the last two or three decades. John F. Ehlers, an electronic engineer turned trader, says, in his book Cybernetic Analysis for Stocks and Futures, “The advantage of digital signal processing offers the advantage of viewing old problems from a new perspective.” He goes on to say that he has developed new trading tools and that: “Traders who master these concepts will find themselves at a great advantage when approaching the volatile market of the twenty first century.” Ehlers discusses trading lag and inaccurate signals generated by the inferiority of preceding indicators to his own designed tools. Here’s a link to some of his papers (warning: they do have a lot of maths in them): http://www.mesasoftware.com/technicalpapers.htm.

There’s someone talking my kind of talk. But can he cut the mustard, walk the walk, stand tall against the Man, to mix cliches and movie catchphrases?

In the book Ehlers introduces nearly two dozen new indicators, helpfully including the source code for them too. Unfortunately these are not coded for MT4. However they have been ported over to MT4 by MQLsoft and are available here:Custom Indicators | MQLsoft - [mql, programming, programmer, service, coding, mql4, ea, forex, trading, automated, metatrader, indicator, expert advisor, ea, strategy, ea programming, robot, custom indicators, algorithmic, professional, free, downloa.

I have been using a version of the Fisher Transform and some of the others and I think they show promising possibilities. There are a lot of Ehlers MT4-ported indicators by MQLsoft and others, and also interesting are indicators made by Jurik Research, which claims that it is using programming skills and predictive algorhythms developed for the military during the cold war and applying them to financial trading. I am using a MT4-ported (found on a forum) Jurik Moving Average (JMA) and it does seem to be very trend-sensitive and have less lag than the standard ones. Their four replacements for common indicators are suggested to be faster and more responsive than the older indicators. This echoes Ehlers’ thinking on the subject.

At the end of this essay are links to some of the sites which I think might be useful; of course there are bound to be other advanced indicators that have not come to my notice.

However I’m not capable on my own of testing and judging them correctly. In any case one person’s opinion is hardly scientific. FMP members in general are co-operative, helpful, team players. So I have a project to propose:

The Project

We collectively identify and test the best say 8 – 10 advanced indicators and then shortlist the 3 or 4 out of them and devise a trading strategy to use these indicators in the very volatile times we face for the forseeable future. We only need to gain a few percent greater accuracy in identifying significant forex transition points, such as trend reversals, to make a considerable degree of financial gain collectively. It’s in all our interests to pool our skills and knowledge using demo accounts to test the software. Then live test them. I’d be willing to take this option and I expect other forum members would make this commitment too.

My feeling would be that the simplest strategy would be the most useful. 3 or 4 indicators: trend, momentum, volume, reversal are all I suggest we need in addition to the skills we have already.

We can do this in a scientific way. When we have the shortlist, we can use the MT4 editor to take the names off the indicators and call them, say A, B, C, D. While this won’t totally remove “experimenter bias” – the tendency for experimenters to have their own prejudgements which can skew results, it will mean that this is reduced. We can swap indicators around among the group to see if good results are replicable by other people. This is standard scientific procedure: results must be replicable, and not reliant on the skill or knowledge of specific individuals.

We can post reports on this project in the Members Area of the forum. A few of us have already used Skype to have forex discussions so we can teleconference when necessary.

Continued in Part 2

[marc walton]

Part 2 Photography & Forex

Porting Custom Indicators

The Jurik MT4 indicators are not going to be available until February at the earliest. Perhaps we can collectively encourage Jurik and his colleagues complete their work before too long if they know there is a ready market for MT4 versions out there. This also counts forhttp://emini-watch.com’s indicators the Better Volume Indicator and Hilbert Sine Wave Indicator (which is also considered under Ehlers’ analysis). Maybe if computer-savvy individuals here could offer to port the tools and then receive royalties from the sales. Perhaps FMP members get them at a discount? This would be an added incentive for membership.

We have to be careful to select indicators that are created by technically competent programmers; because the Ehlers source code is available and it seems that MQL (the MT4 indicator code) is an easy language to program in, there are many free indicators around. I have personally tried out several versions of Ehlers’ StochasticRSI indicator downloaded from various websites and I experienced problems and could not be sure of the functionality of the software – that’s why you need professionals like MQLsoft or Jurik rather than individuals to code reliably and test functionality thoroughly before the application is used in live trading.

The Volume Problem

Forex trading does not have volume statistics in the way that stock trading does, so you can’t tell whether there are high volumes of transactions, or in slow periods, few trades. This is something that would be useful, to tell if a movement has slowed because not too many trades are being made, or whether major players are getting out of the market (a sudden reduction in volume) because it is likely to reverse.

In science, there are what are called “proxies” – ways to measure things you cannot see. For example, many planets have been discovered orbiting far-away stars by measuring the fractional dimming of the light produced by those stars, and calculating the size of the planet from that. You may have heard that an Earth-like planet has recently been discovered 600 light years away by NASA’s Kepler Mission, orbiting a star not dissimilar to our own sun. It is probably 2.4 times the diameter of Earth with a year of 289 days, in the “habitable zone”, where water will be liquid and therefore life is possible. Scientists have computed all that without ever having seen the planet directly. For another example, the size of illegal drug consumption, which obviously cannot be market-researched, audited or taxed, can be measured by taking a random sample of banknotes and testing them for cocaine traces.

There must be a way of finding a proxy for volume and creating an indicator which would give us an estimate of this data.

Then, if those with computer skill in the forum can ensure that the indicators can trigger an alert when the requisite trading conditions from our strategy are met, then this would be helpful. I cannot be the only person who finds staring at charts all day boring, and an incentive to take half-baked trades out of a misguided need for stimulation. I also have other work to do, and I am in the lucky position of working for myself. Many of you have regular jobs and so have to do your trading outside your working hours. What’s the point of the computer revolution if you cannot have a jollop of automation to ease you along?

Intellectual Property

We need to be conscious of other peoples’ intellectual property. If we find a good indicator, whether commercial or “amateur”, I strongly suggest we pay to licence it, unless they are in the public domain as Ehlers’ tools and the MQLsoft versions of them are. There are two reasons for this, firstly that people should receive the benefits of their labour; secondly it is an incentive to improve these tools or create new ones.

There are several methods which might facilitate this: we could each chip in a small amount each to have some tools created or ported, or perhaps they could be sold from the various FMP sites to non-members and be a free or discounted benefit of membership – there’s various options, and it will be a matter of negotiation on a case-by-case basis. According to their website the four Jurik Indicators for MT4 will be sold as a package. I’m sure Marc and Dean are capable of negotiating a discount should we decide that these are worth performance-testing.

Conclusion

Finally, when I go to photograph something, I do it whenever necessary. I don’t wait for a clear blue cloudless day when the light will be regular. I take the requisite tools, in terms of camera, lenses, lighting or other equipment to do the job. I rarely feel cowed by a photographic challenge.

We should be striving to have similar, accurate tools available to the FMP community. Acknowledging the febrile nature of the forex market, I believe this is possible. This would give us an edge in terms of reliability and replicability of trades. I am not interested in gambling. If I was, it would be much more fun to go to the horse races than stare at a screen full of rectangles for hours on end. What I need is an extra, reliable income stream to cushion myself against the tribulations of global financial troubles, which are going to continue for many years to come. I do not know the financial situation of other FMP members but I would suggest that many of you are in a similar position: facing an uncertain future, and whatever level your forex trading skills are at, using more accurate indicators to improve your scoring rate is in your interest.

Over to you now.


Links:

Cybernetic Analysis for Stocks and Futures by John Ehlers on
Amazon.com
Amazon.co.uk

John Ehlers’ site: MESA SoftwareThis has many mathematical papers on the problems of measuring trading and how indicators can be improved.

MQLsoft
Ehlers’ indicators, and a few of their own, ported for MT4.

Damiani Indicators
This programmer has created quite a few of his own and Ehlers’ type MT4 indicators. They seem to be effective and functional, according to various forums.

Emini Trading | Emini Day Trading | Emini Trading System
Although this trading site is not mainly directed at FX or MT4, some of the custom indicators looked interesting to me.

Jurik Research
A ghastly-looking site. My reverse-psychology take is that the indicators must be good if they can get away with such poor design. Here’s the product guide to the indicatorsProduct Guide Information Note that it is fairly light on trading strategies with these indicators, so here we would have to come up with our own.

JMA indicator (Jurik Moving Average) (not created by Jurik)
uploaded to FMP forum here:
http://www.forex-fxtrader.com/forum/...-share-jma.mq4



And finally
Nasa’s Kepler Mission:
NASA - NASA's Kepler Mission Confirms Its First Planet in Habitable Zone of Sun-like Star

[reynuevo]

Thanks for sharing