Wilder and beyond for the part-time trader

Dear all,

this new thread starts from another very interesting and serious one, the �Trading Systems in 'New Concepts�� founded about one year ago by dpaterso and followed by a group of heroes in these months.

Before starting any discussion I find it necessary making a tribute to one of the most amazing and wonderful guys I ever met on the �cyber space�, and namely to Dale Paterson. In making that I hope it will also result clear why I decided to set up a new discussion forum, and therefore this tribute is also the better introduction to the discussion.

In most cases, in trading books and often in trade blogs, you generally find only �success stories�. The usefulness of such kind of reports is normally quite low, because, as soon as you start putting into practice the �sold wisdom� (even worst when �sold� entails also a monetary disbursement!), you notice how life (= markets) are much more complicate, and how theoretic �stellar� performances, or �handbook-like� setups rarely happen again, after the time your mentor was able to show them on a chart (in the past!).

Dale is not such a teacher. Rather, he more often had the braveness of showing his mistakes and losses, perhaps, than the pride of boasting out his wonderful performance. But, in making that, he also showed: [ul]
[li] an exceptional endurance, and an almost stubborn will of deepening the knowledge and refining reliable and waterproof trading methods
[/li][li] an infinite patience and invincible attitude towards sharing his findings, problems and successes with everybody sincerely committed to undertaking a serious process of research and self-development
[/li][li] the value, and at the end the good results, of the application of rigorous and controlled experimentation of the markets and of the systems.
[/li][/ul]

For this and other, quite a lot of us had the honor of becoming his faithful �cyber friends�, going along, from the beginning or from the midway (as in my case) his long adventure.

Dale has also some defects, which make him so �human�, near, and cute. The first one is that he has an irreducible tendency in attributing to J.J. Wilder Dale Paterson�s merits. What I mean is that he really found many �tricks�, or better �correct interpretation� of Wilder�s findings, so to have them serve better to the trading onto current markets. And these �things� were not actually in the original Wilder�s work, or at least were so cryptically or embryonically laid out there, that it was just thanks to the work (and pain!) of Dale that they were got our, clarified, put into practice. I think that this process must now go on further from Wilder and beyond, and, in agreement with Dale, it is probably time to pursue this discussion in another place, leaving the �closest to Wilder� stuff in the original thread.

The second is that Dale is a tireless and sleepless full-time trader. Some of �his� methods, and for sure his constancy, are not easily available for �us poor part-timers�. This thread would like to be a challenge: is it possible also for the �one-to-two-hours-a-day-trader� to be profitable on the forex? Is it possible with Wilder-like, or Wilder-beyond methods? This is another reason, a sort of �peculiar viewpoint� for starting another discussion. So Dale: you�re warned! It will be a BIIG honor and enrichment having you here: but DO NOT ASK US TO STAY IN FRONT OF THE SCREEN AT 1:00 AM! Clear? LOL, LOL!

The third defect (he says that!) is that sometimes Dale likes �seeing what he wants to see�, instead of �what�s in front of your eyes�. And so, here, we would better like to name things with their nouns, and eventually also �destroying� some pieces of the puzzle. Destroying is sometimes the way for building new things up, and this should be the spirit here. So �criticism�, but in its �scientific, constructive� meaning.

This means that everybody is warmly welcome, but, please, let�s try to �make a job here�, so being serious, constructive and positive. And very respectful for each other. That makes life nice and allows building a good mood. Of course this does not mean not to laugh a bit together, when necessary or useful!

I have quite a lot of �introductory ideas� in mind, but would try to alternate a bit of �philosophy� with a bit of �bread�. And, as time, for �part-timers� is always short, I just shut up, and post a first crumb.

Bye

Fabio

First crumb then�
Please look at this link: 301 Moved Permanently

Here, in the �old� thread, you find the bona-fide correct parabolic SAR indicator for metatrader 4.

There is also a bit of discussion on it, and an invitation to contribute in testing and making corrections.

Why metatrader? Why the parabolic SAR? Is it useful? How does it compare to the other available around?

This and more from here on!

Bye

Fabio

Waw, people, what an enthusiasm here!

Who cares. At least this will be a good place for fixing ideas!

Today I would like to make another “introductory note”. And this concerns: why to use indicators?

I must sincerely express all my appreciation for the many people, here at BP and elsewhere, who argue that in trading it is better looking just at pure price action, and indicators are just lagging and useless additions. People like Nick B are inspiring, and surely very profitable traders and teachers. Other really have a “sixth sense” for charts and movements, and for sure are able to produce superior results.

I have just two problems with such an approach:

First: it requires a lot of time. In the beginning for learning. This is a sort of sport, which requires a lot of practice, and improves with time and training. Then also the practice itself of this kind of trading requires being in front of the screen as long as possible, probably, at given hours in the day. Both these requirements are not best suited for “part-timers”.

Second: I have the impression that most these techniques have a quite high degree of subjectivity, and are difficult to formalize. Nick B often says that “the brain” is the most important “trading system” ever invented. And I fully agree. The only little addition I would propose is that “brain” is able to work in trading not only in “real time”. There is a sort of neural activity that can be applied to trading also “before” or “apart from” trading hours and real time sessions.

This would like to be the spirit of the present discussion. Indicators can be useful tools, if well understoon and still more well applied. They are just instruments for taking decisions. Due to the higher level of formalization possible, they can probably also be a good basis for some degree of automation, something that is likely to attract quite a lot a “part-timer”. It is very likely that indicator-based trading would be less efficient and profitable than “brain and judgment-supervised” action. And this is also a question of fixing targets: I’m not aiming at 50%/month profits.
The bet here is to understand whether indicator-based, a bit automated, part-time fx trading can indeed be at least a bit profitable.

Good: open to discussion and inputs.

But before leaving, as there are quite a lot of impatients around ("No more discussions here! Let’s just trade! — mmm: looking for “free-bread-signals” here? :confused:)
And perhaps for increasing a bit of curiosity and discussion.

Here are the current results on a live demo account in the last days with the still very experimental and evolving system(s) I�m currently working on: (times are GMT+1)

10/02 h13:00 short GBPUSD @1.4810. Closed 13/02 h03:10 @1.4384: +426 p
10/02 h13:01 long  EURGBP @0.8774. Closed 13/02 h07:17 @0.8969: +195 p
10/02 h21:46 short USDCHF @1.1563. Closed 16/02 h15:19 @1.1679: -116 p
13/02 h08:27 long  EURJPY @118.03. Closed 17/02 h01:27 @116.52: -151 p
13:02 h08:52 long  GBPJPY @130.78. Closed 17/02 h09:23 @129.66: -112 p
12/02 h10:30 short EURCAD @1.5952. Closed 17/02 h10:57 @1.5992: -40 p
16/02 h13:58 short EURCHF @1.4857. Closed 18/02 h12:10 @1.4800: +57 p
17/02 h07:25 short GBPUSD @1.4150. Closed 19/02 h02:36 @1.4284: -134 p
12/02 h08:57 short EURUSD @1.2833. Closed 19/02 h06:51 @1.2592: +241 p
Balance: +366 p

The orders you see are all limit orders: I did not stay at the computer in the middle of the night. They were then adjusted, as far as possible, in few minutes during the day. Note the disasters of 17/02, because this is a �part-timer lesson�. I was abroad for my job the whole day and did not have a single chance of grasping at my PC. Several SL hit, and an entry that I would have probably avoided. But, as said, this is just an experiment, and fortunately just demo. I would say, anyway: it�s promising!

Bye

Hello fabio,

Intersting new thread you got going on here. I’ll be following along when I can and with interest.

Cheers
Boca

Thank you, Boca,

nice to have you here. Let’s see what I can manage during the weekend!

Bye

Fabio

So, here we are, people (LOL)!

Today we will speak a bit about the parabolic SAR indicator referenced previously. It�s not a �basic� discussion about this indicator, because around you can find many. Especially, here at BP, there were heroes fighting with it for long time, and also� losing incredible amounts of money with it. Look for the thread �Parabolic SAR � that�s all!� and you�ll find out!

The kind of discussion is: how we can evaluate this and in the future other indicators, in order to use them, or their combinations, for profitably trading. This is only a first attempt to make it, and obviously suggestions, criticisms and corrections are more than welcome.

Parabolic SAR was invented by JW Wilder in �New Concepts in Technical Trading Systems�, issued in the far 1978. You all know the �shape� and the behavior of this indicator when overlaid on an instrument chart.

Now: what are the features of this indicator?

First: I would say this is a [B]DIGITAL[/B] indicator. And the opposite would be �analogic�. What do I mean? The signal delivered by this indicator is absolutely univocal: go long or go short, nothing in between. With other, more �continuous� indicators, there will be sometimes the problem of fixing a threshold, or to �interpret� their indications based on some other variables. This is not the case of pSAR. When dots are �below� the price line, this is a long indication, when they�re above, that�s short. Fullstop.

Is this a [B]SENSITIVE[/B] indicator? For sensitive I mean able to rapidly �feel and report� changes of the market conditions. I fear that the major problem of pSAR is to be OVERLY sensitive, or �hyper-reactive�. Anybody looking at it carefully over some time and several instruments will immediately see how many �whipsaws� the pSAR is subject to.

Is this a [B]RIGID[/B] indicator? For rigid here I mean the number of parameters used for calculating it, how variable they can be and how much they affect its output. Well, perhaps this is not a nice wording, because I would bet that most of us would like more �flexible� things than �rigid� (please feel free to suggest a better wording!!!). But in the context of this discussion �rigid� is actually better than �flexible�. Because, when you have several parameters to optimize for a thing working, at least in trading, you very soon come to the point where you don�t know anymore where you are and what you should do now� Overoptimization is always at the door, and the result is that, after a cumbersome process, you come out with something that is working until� yesterday, but will never work tomorrow. In this sense, yes, pSAR is quite rigid, and this is good! The two only parameters, acceleration factor and maximum acceleration were already extensively tested by Wilder himself, and can be only very slightly changed (though probably with horrible results). I would suggest keeping them equal to 0.02 and 0.2 respectively. All the rest, this my impression, would just get things worse, or be an �overfitting� (improve indicator�s behavior on a particular instance, but impair it on the most others).

[B]What is this indicator good for?[/B]

Well, this question is much one of the topics of my current research, and I don�t want to get any answer here. The only thing I want to mention is the paramount importance, if using pSAR as it is, of the �[B]significant � or extreme- point rule[/B]. This is one of Wilder�s �gems� (in my opinion more than many others), and its connection to pSAR is NOT Wilder�s discovery directly, but much more of our super-friend Dale. The instructions for the �Parabolic Time/Price System�, in Wilder�s book (the one where the indicator and its application are introduced) are actually very misleading, and it was just Dale�s merit having recognized the importance of the extreme / significant point for pSAR use.

This means: NEVER EVER enter a position �straight at market when pSAR signals a reversal�. This will cost you your shirt! Just place a limit order at the significant level the price reached in the opposite direction. I.e.: when you want to enter long, because the first SAR dot has appeared under the price line, then put a limit (or stop, depending on the naming of these orders at your broker) buy order at the high of the bar in the previous bearish trend which is the closest to the turning point. Go backwards, from the current bar into the previous downtrend, till you get a bar surrounded, on the left and on the right, by bars with smaller highs. Use the high of this �significant bar� as your entry level. Be assured: if you make that, you’ll be saved from an incredible number of very bad experiences!

For today, not to write a novel, I�m over.

Hope in stimulating discussions and deepening wisdom!

Bye

Fabio[B][/B]

Hey Fabio!!!

EXCELLENT THREAD (and I’m PARTICULARLY fond of the first post)!!! LOL!!! (Your check is in the mail)!!! (On the other hand: being THAT nice to me / about me COULD make you VERY unpopular you know)!!! LOL!!!

Listen up: I wish you ALL the ‘luck’ (I don’t believe in ‘luck’ in this business but you know what I mean) with this thread. I shall ALWAYS ‘pop in’ here (I only found out this morning that you’d actually started another thread). I’m not sure how much (if anything) that I’d be able to ACTUALLY contribute here because I’ve a feeling that anything that I MAY contribute here would end up being a repeat of most of my ‘inane drivel’ on the other thread.

That said: I’d say to check BOTH threads from time to time (as an example I’m about to post yet another of those posts ODDLY ENOUGH relating to what we / I ‘WANT to see’ as opposed to ‘WHAT ACTUALLY IS’)!!!

Regards,

Dale. (forexbrokersonline.net).

Edit:

ACTUALLY Fabio: you’re always talking about your options trading. How about IN LAYMANS TERMS you explain how they work? I for one DO NOT QUITE understand the ‘logic’ behind them (no matter HOW many articles I’ve read about them) but I’m TOLD or rather LED TO BELIEVE that options are THE things to be trading???

Delta has gone to great length to explain currency options here:

Deltastock AD - Speculating with Currency Options

and here:

Deltastock AD � Options and Terminology

but I STILL ‘don’t get it’!!!

Hallo Dale,
it’s a honour having you here, and I never cared in life, nor I’ll start now, being “popular”.

I’m also switching to your (“our”) original thread forth and back. The reason for having this new one was to keep here “experiments” also “beyond” Wilder. You’ll see. Let’s say that, with some more “participation” I will be “more strongly motivated” in keeping up faster. Just an anticipation: what is now in my mind is:
[ul]
[li] some still introductory note on why and how metatrader
[/li][li] some comparative note between pSAR and ADX (or better: the DMS)
[/li][li] the introduction of other indicators, which can complement the weaknesses and eventually exploit the strenghts of Wilder’s. First topic will probably be the Schaff Trend Cycle
[/li][/ul]
All will be done, possibly, by building up (hope collaboratively!!!) a trading system and methodically (back-)testing it, by keeping an ordered record of pitfalls, improvements and parameters.

Sorry if I cannot write much more now: you know, for me writing here, in addition to being a great opportunity for learning from you all, is also an opportunity for “fixing” “ordering”, then “making sense” out of sparse ideas, episodic findings and punctual observations.

But I think that, already writing down a sort of the “index” I presented above is hepful for me and hopefully will stimulate somebody else start looking around (and putting out ideas).

Concerning options: thanks for the links. I don’t know options on currencies, yet, and it will be a pleasure looking at them. What I use are options on stocks. Again: very keen to speak about them. The only question is whether this is the correct place, being a forex portal. And, further: you know, there is so much to do and to speak about… it’s a matter of priorities given limited resources!

Anyway… let’s try going on. Now I’m really curious about the last news in the “main” thread, and switch to that.

Bye, and to next time!

Fabio

Dear all,

here the (hopefully) last �introductory post� before starting the real work.

So: why metatrader?

Metatrader is in principle a good piece of software, with nice charting features, a quite good tester and a well integrated programming language. Again: it is very popular, and a lot of brokers (even Delta!) offer it, usually also in free demo accounts. As far as I have understood, it was developed by a group of very skilled IT professionals of Eastern Europe, who maintain a very nice web site, called the MQL community (MQL is the programming language used within MT).

There are anyway quite a lot of bad features. The first, and most serious for most �common users� of the platform, is that many �standard� or �custom� indicators are completely wrong. MT is released with a standard collection of TA indicators. For some reason, some of them are included and classified as �standard� and other �custom�. Of these latter it is possible to see (and modify) the source code, while for the first series just the compiled code is provided, meaning that you cannot modify them (at least not easily�).

Well, at least for the Wilder�s ones, there is NONE of these �delivered� indicators which really corresponds to what JW Wilder was actually writing in his book! I must say I did not check the RSI, but for parabolic SAR, the ATR and the ADX: none of them is correct! And the differences are notable and serious. We�ll probably have the chance, here, to look at some of them. This is due probably to the nature of the developers, coming more from the IT than the finance world� And this is also the source of a lot of misunderstandings about for instance �Wilder-style trading�. People try to use �MT-ADX� for instance, sees that it does not work, and concludes that Wilder-trading does not work. Well, perhaps we could come to the same conclusion, but I feel it is serious doing that just after having deeply investigated what �Wilder-style� really entails, and not the �wrong version of what is supposed to be Wilder�s�, right?

Besides that, there are other minor problems: the tester is far from being perfect. The language is poorly object-compliant, there is no debugger� and so, sometimes life with MT4 is quite cumbersome! But, nevertheless, the very nice feature of the program is that it is flexible, and you can get it doing (almost) all what you want. For this reason, it is the freely available tool for testing and experimenting with trading systems, in my view. No compiler needed, no API to deal with� Just keep the �box�, discard some of its contents, and make what you want!

Now, why I think it is so precious and important using it for testing?
I have the impression that any observation we can make on markets and respectively on indicators is always �episodic�. We see in ten-fifteen instances the good (or bad) interaction of price action and indicators. And then we get convinced that this interaction is extensible on the generalized behavior of markets. An example of that is the beginning of Dale�s �pSAR-that�s all!� thread, where the initial wonderful results were then denied (in the painful way) by the future dismal events.

Unless you have really a lot of time (more than the average �regular job� working schedule) an obvious solution to this problem is backtesting �by hand� years of data, observing these interaction as they happened in a lot of instances in the past. This must be performed �the right way�, so, looking really bar after bar, scrolling them one at a time from the right side of the screen, and taking written note of the observations. This must ideally be performed also on several different instruments, to have an idea of how general the observation can be considered. The discipline required is quite robust, because, if you don�t want to cheat yourself, you must avoid introducing biases and �tricks�, and this becomes more and more difficult as the test goes on, because with the �experience� you might start understanding that the original system you are mentally testing has to be improved by this or the other modification. And so you end up not testing the system, but the evolving concept of that. And eventually it is also possible that, the new improvements would have caused a lot of problems in the already tested past period. And nobody knows whether future will look like mostly like the recent past or the far past, or something completely different again� Besides these �theoretic� problems of manually testing, well, this is not for the �part-timer�, and, in this thread, this is enough!

What I find great of MT, is therefore the possibility of a fast, neutral, mechanic testing of a precisely defined system over long periods of time and a lot of different instruments. And the possibility of introducing tricks, modifications, parameters in a very controlled �non-cheating� way. Still the work to be done is big, especially if there are several tricks or modifications to be introduced. And the risk is to become �devoted� to this tool and possibilities, so that you lose yourself in over-analyzing and over-optimizing things, forgetting the general rule that probably simpler is better (and overoptimized systems on the basis of past data will probably fail in the present-future because they will be �born old�, adapted to the past).
This place here will not be the source of �free and easy to get EA�s�. Yes, for backtesting with MT4 you must have an EA. I want to stress that my aim in making them is above all for exploiting these analytical possibilities of the software. If then they work, they can also be adapted for the �part-timer� semi-automatic trading, but, their first aim is another (besides that automatic trading �knowing what you�re doing� is quite different from �blind black-box-like!). The second reason I�m making them is that they force you formalizing, fixing, clarifying the rules of the system, without doubts, tweaks, feelings, uncertainties.

Personally (others here are of course free to make otherwise) I will not post here my code or my scripts. I will anyway be very glad to share them, via private, confirmed email, to all my �good friends�. Either the ones I already �know� from other common experiences, or to anybody who will demonstrate to be one by actively contributing for a while and sincere efforts to this common work. The deal should be that these scripts will not be distributed even to your brother or mother, but used for contributing to the common understanding of things here, and to the progress of a trading system. There are several reasons for my choice, but eventually we can discuss it in the future.

The way for contributing can vary depending on personal skills. Of course it would be nice having somebody able to code, and improve on the common platform (and correct my many mistakes!! LOL). But also for the �non-programmer� there could be quite a lot of help to bring, by testing, reporting observations and suggestions for better features/methods. Of course you must have the desire of learning at least how MT4 basically works and especially how to perform tests. Much better (but not required) if you also would like starting to at least understand what is written within a piece of code.

This is an offer. Of course, for the others who just want to take part to the discussion of indicators, systems and markets, there is always a lot of room open, also without MT4, scripts and backtesting!

Over!

Bye

Fabio

Well people,
I say “let’s start”, but given the increasing enthusiasm I must also say that I don’t know whether I’ll go on spending time for putting all together: my time is probably best used by testing… Anyway. At least I have a written record of experiments.

In the attached figure you have the description of the script I’m going to comment. I explain here what it’s doing, and if WE go on together, I will then post the results of some test hoping that somebody else will catch up and add something to it.

What’s the script doing? That is: how does the trading system work? (Just not to give the impression that this is a place for “script freaks”: we are using them, but we are discussing about trading systems and indicators here: it’s good also for the ones who want to go on by hands!!!)

  1. High and low significant points. To determine them, and this is an interpretation of Wilder’s logic (so, please, think about, and if you find it wrong, let us all know!) I implemented a recursive algorithm making the following (explained for Low sign. points, for HSP is the reverse):

take the last closed bar and look at its low
this becomes the current LSP (low signigficant point)
[ul]
[li] if it’s LOWER than the previously determined LSP
[/li][li] OR if it’s lower than the previous bar’s low.
[/li][/ul]

Look at the bottom left of the figure (EURUSD, Daily, Feb 2008). The first LSP is indicated, and it’s quite obvious. Then the algorithm, every day, checks whether the new low is either lower of this LSP or of the previous bar. This way the next LSP gets fixed on feb 18 (and you will see how this is there also evidenced by the new red arrows).

The converse is true for HSP: change it to a new one when this is higher than the previous HSP or higher than previous bar’s high (there is unfortunately no such a “visible” example in the screenshot presented).

Second: what is the system doing?
The system operates just after the opening of a new bar (start of the new trading day). The pSAR dot is calculated and drawn on the chart in that moment.

As soon as a new parabolic SAR signal comes (short or long: meaning pSAR gets under the price line or respectively above), the system places a stop sell/buy order at the level of the current LSP/HSP.

I coded the system so that, if the stop order is not hit till the end of the day, at the new bar it is immediately cancelled, and eventually, if the pSAR has not changed direction, replaced. This way what I am achieving is something like “while the SAR is in a certain direction wait until price hits the significant point”. As a corollary this means that actually I’m not trading just when SAR “reverses”. But instead I place orders in the current direction of the pSAR: if pSAR < prices, always try to set a buy order; if pSAR > prices, a sell one. I’m very aware that another variant (something for next testing) could be: try to enter just when pSAR reverses. If it doesn’t work stay out and wait for the next turn!

Another consequence of my way of coding this system I’m well aware of is that, in doing so, if the SP changes while still in one direction but not yet in a trade, the limit of the orders will change too, as it is the case on 18 Feb (and you now know why there are a series of red arrows).

Once in a trade, every day use pSAR as a trailing stop: move SL levels at the level of pSAR until the trade is closed by hitting the stop. At that time get your gains (or losses!) and start the cycle again.

This is the system. And it’s open to comments and suggestions. What you see on the chart is very easy: a red(blue) arrow indicates a short(long) order. If it’s not executed the tester stops it and places a green arrowhead. If it’s executed you can follow how long it goes on by following the dashed line (and of course you easily discriminate the winners from the losers) till it’s closed, always by a green arrowhead.

It’s easy to make (or re-think) the observations about the behaviour of pSAR that were previously discussed.

The quite amazing thing, but this is just an anticipation of a more serious analysis, is that, surprisingly, in 2008 this “dirty” system was profitable. But this probably was due to the abnormous downtrend due to macroeconomic conditions. Normally (in other years in the past) the system is sucking. And this is even more true for shorter timeframes.

In the next posts (if there will be some) we could start analyzing ways (if there are such… but I have some idea and already some experiment done) for improving this system or finding better ones!

Bye

Fabio


Fabio,

I too am very interested in utilising an accurate Wilders Psar, particularly for closing rather than entry.

I currently do all my backtesting in excel and prefer to use several columns to produce an indicator rather than use a macro function. This allows tweaking of the indicator having once tuned it to reflect as acurateley as possible the author’s intentions.

I am really struggling with excel formulas to produce a Psar indicator. The only working one I have is the Ta-Lib macro which I cannot alter as source VBA code is not released and also it is limited to first 65000 lines of spreadsheet which is a severe limitation. Currently I am backtesting on 7 years of FTSE index and 10 years of EUR/USD Forex, both in 10 min time frame.

I note from your contribution to the Wilder thread started by Dale (which I have read with interest) that you have generated the excel formulas to approximate a Psar indicator for comparison with the MT4 built in indicator. I would be extremely grateful if you would post an actual excel sheet with the formulas in.

Hopefully when I am able to introduce a Psar indicator in to my back testing I will be able to make a more meaningfull contribution to your thread. I think my backtesting is sufficiently unique to be of interest. As well as testing various hypothesis either read about or dreamt up; I often get interesting results from “black box testing” that is to generate random variables in excel and let repeated calcs take place overnight recording the best results.

Thank you for your interesting posts to date

Leo

WOW, Leo, as my friend Dale said me once upon a time: A BLAST FROM THE PAST!

You might have noticed two things:
[ol]
[li]There has been no activity on this thread since ages, and, actually, it never “started up”, really! LOL.
[/li][li]There has been no activity from my side on this entire forum since ages, now!
[/li][/ol]
This must be explained somehow. I actually decided for myself that I am no longer posting on such public blogs, and especially on babypips, and this reply to you now, is an exception that I’m eager to accept because of your kindness and sincere interest.
All depends on a very bad situation, some months ago, when a nice guy was starting “explaining” (actually: disclosing a bit, in quite “exoteric chunks”) some very interesting ideas about “market wisdom” (connected with mathematic patterns recognition) claiming they were very successful, in his hands. A horde of barbaric envious veterans of this babypips forum jumped in, destroying any conversation before it could take over. Some of us, interested and “democratic” readers tried to complain against this attack, and also solicitated moderators taking some action. The action undertaken by the moderators was deleting [I]our[/I] complaints and sending personal messages of warning, letting the horde going on till the diclosing guy decided it was a waste of time trying to help people (who didn’t want to hear) understanding his “secrets” and decided to disappear.
I decided, in that situation, that this cyberspace was useless for me too.

I have now a restricted-access personal web site, shared with other friends, mostly known here around, where we share knowledge, experiences (both nice and bad), observations, and trading. I can anyway tell you that, after having started as “straight directional (mostly forex) traders”, going back to my previous experience (previous to forex and babypips), we are quite strongly shifting, nowadays, to options trading, mostly delta-neutral and pure probability based (so, without much interest for the direction of the market, saying it in a very simplified way). I am not going to disclose our experience, nor grant access to this site to the public, because I was really annoyed by such parasites met here in that case. I make it just for people that I trust after having shared a lot of efforts and reciprocal esteem, as for some of the co-authors of Dale’s “epic” threads (Dale, of course, is one of these friends).

Concerning the pSAR excel, I have two problems. First: :rolleyes: I must look for it…! Who knows in which directory of one of the three PC’s I daily use it has been forgotten! LOL.
Second: I will NOT post it here, anyway. It is not rocket science, of course. Everybody can make it. But it is work: and somebody is sometimes not so keen sweating a bit. And I am not disposed making discounts to the horde. I am no way speaking about you, and, if we find another way (and I find the file! LOL), I will be glad sharing it with you, if you promise not to disclose it publicly. Suggest me a way.

But, finally: I will not suggest you to use pSAR much, actually. Better said: I would strongly suggest you exploring non-directional, hedged derivatives trading (…mmm: I should say [I]options[/I], because e.g. futures are quite different!) as a much better, safer, stress-less, “intelligent” (you can “calculate” your risk/profit profile much better) way of investing/making money.
If you want to stay “directional” (and, to a certain extent, of course it would help also with options), I have mixed experiences with pSAR. A suggestion would be: chose a reasonable start point for the calculation (not the Wilder’s “standard” ep, but instead a stop loss level making sense to your trade when you open it) and then calculate the trailing stops with Wilder’s algorithm but starting from that point. I made an indicator for Metatrader4 called “parabolic STOP” (where I decided to take away the words “and reverse”) working exactly like that.
In comparison to that, I had better results with another indicator, called MTEI_Supertrend (or similar), as a trailing stop. I found it around here and in other web sites, and had to modify it for MT4 because it was a terrible re-painter.

But you know: it is since the first days of this year that I even do not open an MT4 terminal, and so, all this stuff is getting a bit dusty. All dates back to the times I was trying to capture the secret of market waves (“oscillatory behavior”?) and extract some mathematical pattern with some reliability. The experience was always: tremendous streams of wonderful gains… followed by nice and plain streams of losses eating the whole away. Even with any scientific and reasoned and proven risk management (remember: risk management reduces the speed of ruin even to almost zero, but cannot make you rich, alone). So, I decided (and I think, definitively), to go back to my beloved options, where profit potential and consequent risk are attached to a defined probability [I]before[/I] you enter a position and till it ends.

Good luck, and let me know if I can help you further!

Bye

Fabio

Fabio,

Thank you for taking the time to respond, I think I missread the year on your post otherwise I would not have bothered you.

I was sadend to learn the reason you no longer post on public boards. I am a noob to the world of the financial markets having studied for about 18 months now and just started posting here and on trade2win. I am continually surprised at how many people feel the need to insult others and their ideas, much better in my view that we respect each others ideas no matter how ill founded they may appear and help each other along. Any pretenders and boasters soon appear to show themselves up they do not need any help from others.

I note your advice about delta neutral options. This is on my list of things to study in the future. I like the idea of not being direction dependant and it is the calculation and cosideration of probabilities that as a Chartered Electrical Engineer I studied at degree level, that sparked my initial interest in the markets. At this stage I have little to offer so will not ask to join your private email groups.

However before options, I must satisfy myself that their is no system applicable to the volatile forex and index markets that is based on pure mathematics without any kind of judgements or inputs from fundamentals.

To this end I will continue with my excel models which i like to think are not strictly direction dependant. My starting model for example is simply deciding upon a number of bets per day and then opening at a random time and closing after a random period. You may not be surprised to learn that applying this too the FTSE index for example with a spread of 2, the cosistent result for say 4000 bets over 7 years is a loss of exactly 2 points per bet. Rather dull and elementary I know but I think a usefull excersice as it proves the market moves randomly from any given point and that my model is not curve fitting which is important as they get more complex. With this info and analysis such as the average points moved for each individual minute of the day over 10 years, by applying some very basic entry and exit criteria along with time of day criteria I am able to produce models that backtest breakeven or small profit. Currently now getting a little more complex with the adition of ADX Bollingers and hopefully soon Psar to tune entry/exits.

If i can work out how to send a private email I will in case you come across your previous PSAr work in excel. As you say I may be better working a psuedo Psar without the switch, also as you point out it is the starting parameters for this recursive calc that confuse me:confused:. I know what you mean about finding files, I run my own Consultancy and have five pc’s networked with server and backup, one must be very disciplined in storage and backup if you want to find last weeks work never mind last years!.

May I finish by wishing you and your group every success in your pursuit of the holey grail of options sytems

Kind Regards

Leo

PSAR dots are a great way to deplete the equity in your account!
Don’t believe it?
Try them.

Thanks for your kind words, Leo.
Your backtesting strategy seems very sound.
I also have a scientific background (…mmm and this is a difference: genetics instead of engineering… and this is a huge one).

finally… I think I have already replied in my last to Leo. I am quite convinced that [I]every[/I] thing applied just mechanically either is unprofitable or has very low profit/risk profile. Wilder’s idea of “technical systems” (which we could rephrase “mecahnical systems”) is, in my view, a bit weird [I]in the long run[/I], because markets change.

If you want to have a good “scientific” reference to what I mean, just google “marketsci” and have a nice (long and cumbersome) reading of the “quants” community. It is hard, but a very “educational” approach.

Oh, finally. Leo: sorry. No time to dig into directories, yet!

Bye

Fab

One person came by, to this thread which seems hidden in the byways of babypips… years later, me. … on the trail of Schaff indicator info.

thanks for all the information , you help me a lot.