I don’t know anything about how commodities/futures are traded. But for options, “cycle” has a very precise meaning. It’s the expiration cycle. And this corresponds “basically” to this average 22 working days.
Could it be something similar for the instruments Wilder was (is) trading? But, then, he chose the 14 days period (including non-working/trading days). Eventually, always for options, also these days are producing their effects (e.g. on the “theta” variable). No idea if it holds true for commodities too (you are the experts). How then this applies to forex is for me still more complicate…
I think that is VALUABLE insight. Thanks for that.
To the best of my knowledge: ALL of these things (commodities, options, and futures including forex futures) have contract expiry dates so maybe this is something to look in to.
At the moment I’m going through ‘chart by chart’ with ADX(11) to see if there are noteable differences between ADX(11) and ADX(14) or ADX(7) and, although in most cases the differences are REAL subtle it would APPEAR that no matter HOW subtle they are there is indeed a difference. Only J.'s automated backtesting over a longer period of time will confirm though.
Of course: if there IS INDEED a noteable difference then this COULD affect something like the VS as well i.e. instead of using ATR(7) maybe we should be looking at something like ATR(5) instead which, of course, it would make it ‘quicker’ to react to price changes. Also: we have 5 bars per week so this makes ‘sense’ or should I say it ‘feels comfortable’. I’m not sure at this point though.
ALL of the above being said though: all my open positions (losses) are turning again so let’s wait and see what the final outcome is of all of this. I DID NOT place any stops or SAR orders last night by the way although as far as I can tell VERY few of them would have been executed anyway which is a very good sign!!!
By the way: if we WERE ‘the experts’ we’d not be having this convsersation!!! LOL!!!
While I’m going through the charts this particular one (attached) of AUD/CAD always ‘bugs’ me.
Take a look at the chart and you’ll see what I’m ‘on about’.
There can be NO question that the pair was trading in a range during the period that I’ve drawn the ‘blue box’ around. ADX(14) does not pick up on this until the pair has all but broken out of the trading range. ADX(7) obviously reacts muck quicker but, as we’re finding out at the moment, ADX(7) is far from ‘ideal’ over an extended period of time (I’m using ADX(11) on this chart by the way). What IS NOW very interesting is this: as soon as the pair started trading in a range just look where the DX was i.e. it fell below 20 and simply stayed there until the pair broke out of the range. COULD THERE BE SOME MERIT in applying the same rules to the DX as we have applied to ADX i.e. if DX is below BOTH the DI’s. That said: the DI’s are also an ‘average’ so maybe I need to dissect the equation further???
Just stop kidding! If you all here are NOT experts, who other?
BTW: is it not “risky” shortening “time frames” in trading [U]and[/U] indicators? This looks like a common theme here in the thread, and several times you came back afterwards “repenting” for having done that…
…and another thought: isn’t it just better to “miss” some “wonderful trades” (i.e. actually maybe getting them a little later) but stay away from bad adventures then try to get all of them at the risk of getting also big losers? I have the impression that this is always a tradeoff. I mean: in post #1158 you were saying you were no more avoiding trades with ADXR<20 (actually you suggested to “jump in” there, to get the breakout). But for instance in #1103 you said that, in the “attitudes for positive trades” vs. the ones for bad trades, there was also “checking ADXR daily” (sorry I’m just citing “by heart”). I mean: I don’t want to make an issue now on ADXR. But I feel there should be something here that is going to point at the “trending quality” of these trades with the DSM. If ADXR or similar (VHF? CSI?) is keeping you out of what, in retrospective, looks like a wonderful trend (#1163 on how easy is to see the trends “afterwards”, but not in the moment they’re starting showing up), with the plenty of opportunities that these systems are generating [I]daily[/I] you don’t need to “let your capital on hold” due to this “Trend-indicator restraint”, but just go for other instruments. Or NOT (sorry: I know I’m almost an idiot, yet ).
And, definitely, IF the trend is a trend, I am sure the “indicators” (your good ones, now I would say “coming FROM Wilder”) will also get you in…
This could imply that you’re not chasing the “5000 pips trend in two months without interruption”, but maybe that you will total the same 5000 in five or six shorter trades in five or six instruments: who cares?
Again: some interesting and very ‘apt’ observations. Just one thing though: you must remember that whatever posts you’re reading of mine / ours were posted LONG before this ‘bombshell’ was dropped. J.'s backtesting of the DMS showed that over periods of YEARS there were NETT LOSSES and those figures just cannot simply be overlooked because it SUITS US to overlook them now. In isolation the DMS made an ABSOLUTE FORTUNE over the past eighteen months or so but what happend BEFORE then, pretty much since our daily charts begin, was LESS than ‘stellar’. I don’t know about you but I don’t have the time nor the capital and CERTAINLY not the ‘patience and restraint’ to sustain three or four or five years of consecutive losses in the HOPE that things are going to turn out fine eventually!!! SOMEWHERE here there is a ‘golden set of parameters’ for the DMS. I firmly believe that we’re ‘close’ to finding them. There is, of course, an alternative to all of this: trading the WEEKLY time frame with the DMS. As far as I can tell it makes a HUGE difference especially in these ‘choppy’ times. Whether or not it’s a ‘sustainable’ option i.e. it will take LOADS of ‘patience and restraint’. I mean: I’ll wager that it’s difficult enough for all of us to be trading the daily time frames because of the ‘lack of action’ and that’s why I make the comment about whether or not it’d be ‘sustainable’.
Now: as I go through the charts again I’m noticing that an alarming proportion of the current ‘losers’ were opened when ADX(14) ‘fired off’ a signal BUT ADX was, at the time, falling (even although the DX was rising at the time). AGAIN: just an observation.
(Thanks Fabio: it’s your post that got me to thinking)!!!
Quite simple really:
Most contracts and options ‘expire’ on the second or third Friday of each month (not 100% sure WHICH Friday). Calender days are therefore being used not calendar months. So: there is your 28 days i.e. 7 days x 4 weeks!!! Half of this: 14 days. BUT THEN: if you exclude weekends from this then you have 20 days and half of this is 10 days. ADX(10)??? VS with ATR(5)???
Interestingly enough I’m almost sure that the number ‘10’ ‘features’ in Donchian’s work (and I know for sure that the original formuls used by Keltner for his Keltner Channels used a period of 10)???
Also: 100% of ‘retail traders’ DO NOT HAVE Saturday and Sunday bars to look at.
Also: most commodities are NOT traded 24/7 but only during certain times Monday to Friday i.e. 5 days per week.
All I know is this:
At SOME point we (I, me, anyway) have to ‘pick a period’ and ‘stick with it’ and trade.
I always kind of wondered where the 14 came with… and I assumed that it was based on some kind of ‘week/month’ measurment… But always wondered if that applied to us Forex traders. With luck, J will be able to test some more pairs with more constants, and see if any are better than his previous results. I always thought 10 might work, since that’d be two Forex weeks… or something. I really don’t know.
As for a trading update, things are looking pretty bad. All of my positions were in profit two days ago, and now probably 3/4 of them are in the red. I’m sitting at about -10% for the month right now, and even though I’m confident things will turn around, these are trying times. I probably won’t be checking my account too much today, in the hopes that if I give it some time, things will start going my way again (a watched pot never boils, and all that). Until later, gentlemen.
Most contracts and options ‘expire’ on the second or third Friday of each month (not 100% sure WHICH Friday)
For standard options this is the third Friday (1000% sure!), except for some particular ones (indexes), in which case it can be the third Thursday. Actually the “official” expiration is the Saturday morning (market closed) after “expiration Friday”, i.e. this is the point when contracts are settled.
The markets are closed over weekend. But, as already stated, most parameters (especially the ones connected with the “time decay”) go on their way (=change their values) also during “idle times”.
As far as I know there are a lot of “magic numbers” around for indicators. E.g. the Ichimoku system is based on “exotic” numbers like 26 periods (it was probably the normal trading month in Japan). Oddly enough, at least in some discussion forum about these methods applied to “western markets” it comes out that changing this parameter to adapt it to “western markets” does not improve the system much, rather sometimes gives bad results. At least in this case people are claiming that the role of the indicator is just to give clues as to how to better “read” (interprete) price action of the underlying instrument… In my hands, thanks to it (with 26 periods) I was able to nicely follow with options the last dowtrend of IBM shares (cashing quite nice $$). Nice $$ that anyway keep me by far and large VERY AWAY from the point of having the capital Dale was speaking about for not taking care of a lot of consecutive losses LOL LOL!!!
Thanks a lot, Dale, for your “warnings and corrections”: I really appreciate it because I’m learning a lot from them. Keep going on!
ADX(10) and VS using ATR(5) i.e. anywhere where a ‘period parameter’ was 14 I’ve changed to 10 and I’m going to see how this ‘pans out’. So far: it’s changed very few of my previous entry points. In some cases, however, it’s made the difference of a position STILL being in profit and now showing a loss. I’m not sure if this is ‘THE MAGIC NUMBER’ (if there is such a thing) but this one at least makes ‘sense’ to me.
I suppose the REAL point is this:
This CONSTANT un UNRELENTING ‘playing around’ with parameters on a ‘minute by minute basis’ causes MORE trouble than using a ‘BAD’ parameter in a system i.e. I’ve been ‘caught up in this trap’ far too many times. You stop concentrating on what you’re doing and the next thing you know you don’t know WHERE you are anymore.
And I agree with you Fabio i.e. relating this to ‘The Delta Phenomenon’: Wilder worked out that Saturday’s and Sunday’s MUST be INCLUDED. That said: I’m not using ‘The Delta Phenomenon’ nor do I trust it. What I DO know is that anybody trading a commodity is looking at the same chart that I’d be looking at to base their trading decisions and guess what: they’ll ALSO only have 5 bars per week in front of them. For THIS reason and this reason alone I’m switching to 10 and staying with it!!! I HAVE to trade and make money. It’s really just as simple as that.
While there are some posts from today asking me to compare the different ADX parameters, Dale’s latest post sums it up just as how I am thinking now. Just go and select your favourite number and use that. It’s better that I don’t now throw any extra oil in the flames around this subject. Don’t forget the last pip syndrome! While a certain slice of historical data may give you the best yield with a certain parameter, another slice may opt for another, and there is no way that could tell us which one of them will work best in the future. We only know that there is a range from within it is justified to select a parameter. Let’s not make this too complicated.
PLEASE do not get me wrong here: the work that you have done is INVALUABLE!!! Were it not for your efforts I GUARANTEE that we’d all be sitting here a few months from now saying ‘but it used to work??? Why is it not working now??? What’s changed???’!!! If NOTHING else it’s made us all ‘wary’ and hopefully this will lead to ‘caution’ on our part!!!
Now that being said: I’m not sure I’d use the term ‘favorite number’ for the choice of period!!! For me anyway: the ‘magic number’ HAS to be based on SOMETHING at least!!! LOL!!! Now I’ve been doing some MORE thinking about all of this and the more I’ve thought about it the more ‘correct’ the number ‘10’ ‘sounds to me’!!! LOL!!! My reasoning is this: as I said earlier the contracts are based on a number of days interval not calendar months (nor lunar months for THAT matter)!!! So: that’s 28 days less weekends and that gives e a ‘full cycle’ of 20 days and therefore a ‘half cycle’ is 10 days!!! Now I’ve ALSO noted that even during ‘idle period’ the ‘count continues’ (as per Fabios’s post and the fact that ‘The Delta Phenomenon’ back up this idea’). BUT: we do NOT have bars for Saturday’s and Sunday’s. What we DO have on a daily chart though is the MOVEMENT between the close on Friday and the open on Sunday represented by the opening gap so we ARE therefore taking this movement into account not so??? (Put another way: I’ve managed to convince MYSELF that ‘10’ is the ‘magic number’)!!! LOL!!!
I’ve had other disturbing thoughts today though as well!!! One of the MOST disturbing is that MAYBE this is JUST the way this business works??? Maybe this has answered my question of before i.e. how did all of these people with YEAR AND YEARS of experience lose so much money over the past eighteen months or so. Maybe they HAD ‘systems’ that have served them well for a number of years and, all of sudden, they stopped working and their losses were the end result. Maybe on another ‘level’ this is indeed how it works i.e. just a natural ‘ebb and flow’ of wealth between (ultimately) individuals???
Maybe our problem partially comes from relying too much on one system. In the same way that we spread our trades out across different pairs to reduce the risk perhaps we should aim to have a couple of VS trades, a few PSAR, the odd FSS and DMS, some TBPS and RTS with maybe even the occasional ADT for luck - all cherry picked using ADX/DX.
Something else I’ve been thinking about but haven’t done anything about (the day job is really interfering with my testing!) is correlation between some of the pairs. I’ve seen it mentioned elsewhere on BPips and other forums but wouldn’t claim to be an expert. I do get the feeling that some pairs seem to move broadly together, so by taking a similar signal on both you’re effectively doubling your exposure good or bad to the next movement of the market.
By the way J. I just want to agree with Dale (could be dangerous) to thank you for all your work in helping to testing ideas - I find your (and the other contributors) insights very helpful in keeping focus on what’s important rather than chasing all over the place.
YUP — then we could start a new thread, based on a cooking show — we can call this system “alphabet soup”, then we could work thru chicken soup, beef barley, tomato and minestrone !
your idea on the pairs that work together is the cats meow, but if you really want to be the owner of the golden goose and an up and coming master of the universe and the space/time continuum, entitled to the mansion, trophy wife, maserati and spoiled kids, simply find one that moves the OPPOSITE direction, and you will have doubled your fun and kept your margin very well covered, unless theres a philosophical reason behind youre NOT shorting one of them ! (funny though it may sound, i have met people who refuse to short because they consider it “sinful” ---- ho hum !
enjoy and trade well
mp
[B][I]Within the great hall at Elfinore stands a wondrous coffer, precisely four cubits square and securely latched against the outside world. Inside that repository, shut away from impertinent eyes, abides many an intriquing trading secret garnered from around the world and over the ages !
As a child, i used to watch from the darkness as the secrets were debated and annotated by the elders. No one there held a single thought of my presence – BUT I KNOW WHERE THEY HID THE KEY !![/B][/I]
First off, MP, welcome back. I always get a kick out of your posts. You’ve been away too long.
Well, things are starting to turn around. I’m back up to -5% from -10%, so that’s an improvement from this morning. I’ve also set some protective stops on positions which are showing the most profit, and we’ve got a SAR signal on EURSEK (I think, I’m not looking at it right now). Now we’ve got to see our system turn things around. I just have to believe that things are working back to profit.
Nope. There is ONE other thread that’s longer: it’s entitled ‘Parabolic SAR - that’s all!!!’ As I’ve said to everyone before: if I made money out of ‘waffling’ I’d be a XILLIONAIRE by now!!!
Cody:
(EVERY TIME I type your name I want to add the ‘Bear’ part on!!!)
Anyway: YES things have turned SLIGHTLY but ‘still no cigar’!!! And to makes matters worse: ‘FA’ is going to happen until the US Job’s Report comes out this afternoon.
By the way: with ADX(10) you do NOT have a SAR signal on EUR/SEK AND you got ‘in’ a day earlier. I’m not sure as to the merit of this YET but I’m SURE we’ll find out over the next couple of days. That being said: for better or for worse my EUR/SEK trade is a PTPS trade so I’ve still got a ways to go before the price is anywhere near the Parabolic SAR SAR. I’ll tell you this much ALSO: the ‘irony or ironies’ is that the ONLY trades that are in a profit right now ARE PTPS TRADES!!! Go figure!!! (Well I SHOULD say that of the five MISERABLE trades showing profit: the two showing the BEST profit are PTPS trades and the others are ‘laggards’ by far)!!!
I’ve not posted about this for a while though but I was thinking AGAIN this morning just how RIDICULOUS things can be sometimes in this business!!! I mean: the ENTIRE population of the USA could be out of a job BAR ONE PERSON and as long as the ‘Analysts’ didn’t know about the ONE PERSON i.e. the jobs report was better ‘than analysts expected’ then the Dow would SOAR!!! LOL!!!
).
