Trading Systems in 'New Concepts In Technical Trading Systems' by J. Welles Wilder

[sirkeen]

Thanks Randont. Its certainly overwhelming at first!

Dale, I believe I read correctly that you had created the indicators for GCI as well?

Regards,
M.

EDIT

Now on post 221, and figured I would retract my question. LOL..

[dpaterso]

Good (Tuesday) morning all!!!

As you can see: I've 'deemed' this a 'very important post'!!!

Something in 'the book' has bee confusing me and concerning me deeply for the past couple of months and I've not quite been able to 'put my finger on it' until now (I think). Read on and tell me if you agree with me!!!

It's regarding the use of ADX and / or ADXR.

The question that's always been at the back of my mind has been: 'WHY is there an ADX AND an ADXR'??? (I KNOW what the difference is betwee the CALCULATIONS themselves BUT WHY TWO indexes)??? What I've been doing up until now is finding instruments or pairs with the highest ADXR value ('highest ADXR value' meaning > 25) AS WELL as an ADX value > 25 to trade using the trend following systems and I've been looking for instruments or pairs with the lowest ADX value ('lowest ADX value' meaning < 25) AS WELL as an ADXR value < 25 to trade using the RTS and TBP.

Now it's been bothering me that, doing it the way I described above, very often I'd get a valid SIS entry signal (for example) BUT would NOT take the trade because EITHER or BOTH the ADX and / or the ADXR had a value of < 25. SO: I AGAIN 'studied' Section IV ('DIRECTIONAL MOVEMENT') and it's sort of 'dawned' on me that EITHER I've been reading things wrong here OR there are certain 'typos' in this section i.e. where references to ADX / ADXR have been 'mixed up'. If this is NOT the case then at very BEST case references to ADX and / or ADXR are sort of 'not in logical order' if that makes ANY sense at all!!!

Let me explain:

Take a look at Page 46 (the paragraph that begins 'In Fig 4.12, the ADX ...'. Note that ADX is being referred to here. Now keep reading. In the very LAST paragraph on Page 46 we all of a sudden have a reference to ADXR!!! Can this be right??? Also: between Pages 47 and 48 there are numerous references to BOTH ADX AND ADXR which APPEAR to be contradictory (or at VERY least by following the text literally you'll be kept out of some good trades). The whole point of what I'm saying is that EVEN IF I'm WRONG about what I'm saying above here then this section is REAL difficult to 'decipher' i.e. WHEN do you use ADX and WHEN do you use ADXR???

THEN:

'Light bulb moment'!!!

Take a look at Page 44 now. Have a read of the paragraph that starts with 'This is all the information we need ...' and MORE IMPORTANTLY the next paragraph that starts with 'These three entities pertain to ...'!!!

Now take a look at Page 40. Have a read of the last paragraph that starts with 'At this point you have probably ...' and pay particular attention to the fact that the words 'ADX Scale' are in bold!!!

Now take a lok at Page 111. Have a read of the paragraph that starts with 'For this reason, the most important index to use ...'.

Now what I've 'got' from ALL of this is the following:

The ADX is used ON IT'S OWN to determine which pairs to trade with which systems IF you're NOT going to use the CSI!!! If you ARE going to use the CSI then it's the ADXR that gets factored in and ADX has nothing at all to do with it!!!

(Important edit:

I now believe the above statement to be inaccurate and incorrect!!! I'm leaving the statement here for continuity however. Please read my later posts for more on the above).

DO YOU AGREE WITH ME???

The point is this:

EITHER references to ADX / ADXR have been 'mixed up' or 'interchanged' erroneously OR because of the constant 'jumping around' between references to them a brain like mine can be SO EASILY be confused!!! I'm not sure whether this qualifies as 'errata' to be posted at the beginning of the thread OR if it's just my brain that can't cope with the 'jumping around' but suffice to say I think I've FINALLY 'got it right' (at least in MY head)!!!

[dpaterso]

Hey 'Sirkeen' (I did not want to 'mix' or 'jumble up' my response to you with the above post. I couldn't 'handle it')!!!

I have indeed created most of the indicators for the systems for GCI's platform. HOWEVER: you cannot even begin to compare GCI's charting module with Delta's i.e. (and don't get me wrong here i.e. I LOVE GCI's platform because it's 'polished' and 'uncluttered' and 'quick' and 'logical') and although GCI's charts are great to look at and 'crystal clear' their scripting language is FAR less 'capable' than Delta's. There is NO way to even get the actual VALUES of the indicators to be displayed e.g. if you want to know what the ASI value is you would have to zoom in FULL on the (my) ASI indicator and position your cursor at the point where you would like to get the value from and then read the value off of the scale at the right of the indicator. This is inaccurate at best. I find also that the indicators are extremely difficult to read i.e. even zooming in FULL only expands them horizontally and not vertically so again: to 'get your bearings' really is impossible. I've thought of using Delta's charts and trading at GCI that way but there is a potential flaw with that scenario because of the difference in timezones. At GCI I use a combination of Excel and my indicators (but probably Excel more so than my indicators) to trade. Also: bear in mind that because you live in Canada you may or may not even be able to open an account at GCI yourself (I know that residents of many countries are excluded from opening accounts at GCI due to various restrictions and / or minimum requirements that have to be met e.g. the USA is one of them). Now I know you're LOATHE to open an account with a broker that is SO far away from you. All I can say to you is this: I've NEVER had a problem with Delta AT ALL EVER!!! Their customer support is fantastic (GCI's is practicallly non-existent BUT I STILL LOVE THEM) and Delta abides VERY strictly to the EU's rules and regulations when it comes to the trading of securities and things so I really don't think you need to be concerned that you're not going to be able to go and 'bash their door down' if you've got a problem of one sort or another!!! Put it this way: even if you opened an account in CANADA the chances of that making ANY difference in the case of a dispute is 'slim' at best!!! The only other thing I can say is that Delta has VERY good reports no matter WHERE on the Internet you look (the same cannot be said for GCI BUT I really DO believe that ALL of the bad reports about GCI have been posted or written by people who have lost money and really don't 'know the game' e.g. I've seen a post where GCI was given a 'scam' rating because their order ON EUR/USD MIND YOU was 'slipped' by THREE PIPS!!! That's TOTALLY unfair. Believe me: if you're trading commodities the chances of you order getting filled at YOUR EXACT price is almost NIL i.e. that's just the way it works)!!!

[dpaterso]

By the way:

I don't normally do this BUT here are some orders (hope you don't miss them).

Don't use the number of lots indicated (I just pulled this from a clients account) i.e. it is YOUR responsiblity to ENSURE that you're not violating our money management rules!!!

The attached orders are all as per the RTS with my 'latest' 'B', 'O', and 'S' sequencing 'offering'.

Also:

GO LONG AT MARKET USD/RUB (RTS but order already executed at the time of posting)

and

GO LONG AT MARKET GBP/JPY (VSTOP System).

Note: TP's will only be set on the RTS positions (if the orders are executed) on Thursday BECAUSE today SHOULD BE a 'B' day for these pairs (well again according to my 'latest' 'B', 'O', and 'S' sequencing 'offering'). With all of these entries the AO and AC are 'right on' (green) and ADX is 'spot on' (< 25)!!!

[sirkeen]

I am not sure that I 100% agree,

On Page 47, in the paragraph that starts, "As a rule of thumb...." I believe Wilder is fairly clear stating that he DOES use the ADXR by itself to determine which system he should use.

Furthermore to that... On Page 111 in the paragraph starting with "For this reason..." he is clear that the ADXR index is primary for trend-following systems.

My understanding is that using the CSI index to rate pairs/commodities to trade will provide the best profit scenario where as the ADXR index provides trading the systems at reduced risk..

Anyone agree....?

Here is something interesting to note...

On Page 48, in the paragraph starting with "Another interesting thing is..." IF we should stop trading when the ADX is below DI+/- why is Wilder opening a position on DMI Chart (page 49) between June 20th, and July 13th, even though the ADX is below the DI+/- lines..... Does that make sense?

Regards,
M.

[dpaterso]

Hello,

Thanks for the input. We're going to have to agree to disagree here OR get some more input from others.

This is MY logic:

Page 40 (last paragraph): '... and that you only trade the top five or six commodities that are high on the ADX Scale. ...'.

Page 44 (bottom half of the page): 'This is all the information we need to follow the Directional Movement Index.'. The NEXT paragraph states (after mentioning the ADXR in this chapter FOR THE VERY FIRST TIME): 'These three entities pertain to the COMMODITY SELECTION INDEX (CS) ...'.

Having noted the above, however, the point you make about the paragraph on Page 48 that starts: 'Another interesting thing is ...' is EXACTLY WHY this section has been bothering me i.e. there have been MANY times in the past couple of months when ADX has been BELOW both +DI AND -DI AND YET the ADXR has been above (sometimes WELL above 25). Now you COULD say to yourself: 'Well OK I'm in a trade and ADX has now dropped below both +DI and -DI so I'd better take profit or exit'. THAT I COULD understand. However: what if (and this HAS happened in the past) you've just been given a valid entry signal by let's say the SIS AND ADXR is greater than 25 so you 'go for it' BUT the ADX is below both +DI and -DI!!! See my problem with this??? And it these tiny 'contradictions' (well I believe them to be contradictions anyway) that make understanding this section more difficult than necessary.

Also (as I mentioned previously)L on Page 46 we're talking about ADX for most of the time then all of a sudden we're talking about ADXR and then on the next page we're back to ADX until JUST before the details of the 'DIRECTIONAL MOVEMENT SYSTEM' are given i.e. we're then back to talking about ADXR and the CSI etc. etc. etc.

One other reason I started to question whether or not I was on the 'right track' was this: take a look at a chart where you can cleary see that the pair is trading in a range. AUD/NZD funnily enough is a 'primo' example of a pair that has been, and is, trading in a range. Now take a look at the value of ADX and then take a look at the value of ADXR. At NO POINT has the value of ADXR dropped below 25 (and has been quite high at some points while this pair has been trading in this range) whereas ADX has been below 25 (even below 20 on occasion) pretty much since the pair started trading in this range. Now this of course begs the question: do I use ADX and / or ADXR to determine whether or not I should be looking to use the SIS or VS or DMS or the RTS or TBPS??? Right now I'd be inclined to use either the RTS or the TBPS on this pair and NOT something like the SIS. That's MY logic (now)!!!

Now just to 'throw a spanner in the works': CONVERSLY if you look at AUD/JPY right now I'd say it's ALSO pretty much trading in a tight range at the moment BUT take a look at ADX and ADXR on THIS pair right now i.e. ADX is ABOVE 25 and ADXR is BELOW 25!!!

I just know this: in constructing my ADX ADXR Daily Work Sheet countless times I've often noticed that I've missed LOADS of trades because I was trying to be 'sure' and use BOTH ADX AND ADXR together 'just in case I was wrong' and you can't do this (obviously). If you wait for BOTH ADX AND ADXR to have values of less than 25 then you're certainly going to miss some 'stellar' RTS or TBPS trades and sometimes you will find that if you're waiting for BOTH ADX AND ADXR to have values greater than 25 then you're also going to miss some 'stellar' SIS, VS (although I'm not sure that any of this has an effect on the VS), or DMS trades.

So: which is it??? THAT is 'the question' my friends!!!

From what I can see NOW and after reading this section AGAIN for the one thousandth time (probably) MY money is on ADX ALONE UNLESS I'm going to incorporate the CSI into my trading to get the 'best bang for buck'!!!

[sirkeen]

Good evening Dale! Hopefully you get this before you head off to sleep.

I do not consider my self even half as qualified as yourself when it comes to Wilder's systems. After reading your post a few more times and the book, I would have to agree with you. I do still believe there is some validity to using the ADXR values by themselves as they're values are indicative of strength, however if we are missing massive opportunities, something else needs to be considered.

I took a look at 14 period ADX and ADXR values on a few pairs and noticed that rarely do they both make it above 25 except in strong trends. However, looking at 7 period ADX and ADXR values, there are many opportunities when both are above 25. When you referred to ADXR values rarely making it above 25 along with the ADX, what time period are you looking at? (pretty sure you have switched completely to 7 period)

I do remember a previous post regarding 7 period vs 14 period ADX values, but can't remember if the same concept was applied to the ADXR. I would assume it must have been, as you would want to compare apples to apples.

If the ADXR is an average of the ADX which is an Average of the DX, are we not just smoothing out the values depending on the indicator used? If so, I would like to think that an ADXR over 25 IN ADDITION to an ADX value over 25 would indicate a strong trend. Therefore, using trend-following strategies when both of these indicators are over 25 is correct. If the ADXR falls below 25 but the ADX remains above 25, I believe you are still correct to use a trend-following strategy, however the weaker trend may affect the outcome of the strategy. This is not saying it won't be profitable, but maybe not AS profitable as one where both indicators are above 25.

I am not 100% confident in my analysis... Nevertheless, I can't wait to hear your response!

Regards,
M.

EDIT..

Maybe today's markets are better suited to 7 period cycles... Certainly things have changed since the days the book was written!

[dpaterso]

Hello again,

Hey: I'd HARDLY say I'm more qualified than anyone else on these systems!!! The only thing I'm sure of is that I've probably read the book cover to cover more times than most!!! It does not necessarily mean I'm right!!!

You are correct: at one stage I DID switch to ADX(7) AND ADXR(7) but then switched back to ADX(14) AND ADXR(14). I cannot remember now WHY I switched back but there SHOULD have been a 'reasonable reason' AT THE TIME anyway!!! Having said that: still today (as I draw up my Daily Work Sheet) do I find myself wondering if using seven days instead of fourteen days 'applies more' to todays markets!!! To be honest: I never did do the comparison of ADX(7) AND ADXR(7) as you have just done i.e. at THAT time I'm pretty sure I was using ONLY ADX(7). (I'm confusing myself here)!!!

It's too late for me to have a look now i.e. I need some 'shuteye' but I'll look at this again tomorrow (later today). Put it this way: IF checking BOTH ADX(7) AND ADXR(7) for values above or below 25 occurs more frequently and there is LESS of this 'one being greater than 25 and the other being less than 25' then maybe we're on 'the right track' again here. I'm CERTAIN that ADX(7) 'reacts' MUCH quicker to a change in trend direction (but I think this was the reason I 'switched back' to ADX(14) i.e. I was getting 'whipsawed' by the SIS if I remember correctly!!! Certainly worth a 're-look' though!!!

[randont]

I haven't had a RTS target hit yet this week. Anyone have any clue as to why the market is so quiet? Is it like this very often?

[kaalilaatikko]

Good morning,

This discussion about ADX/ADXR has come into a perfect time slot for me. I have just completed the first reading of that chapter and started thinking how I could integrate this into use. The remarks in the posts have been most interesting.

Wilder states (p. 45) that ADXR must have a minimum fluctuation even when directional movement changes direction. As it is an average of values that span quite a long period, I suppose that it might also smooth out such shorter-term ranging behaviour that ADX would notice and would be essential for some other system.

ADX, on the other hand (still on p. 45), measures movement around the equilibrium points, and its trend should also reflect changes of direction. I would conclude from this that from ADX and +/-DI it should be much easier to see or forecast when a trend changes direction, whereas ADXR might keep you in a trade too long.

Personally I would be biasing towards using ADX, not ADXR. What comes to Dale's "VIP" "Light Bulb" conclusion, I would be one to agree. I might extend that by the following: If you use CSI, use ADXR for ranking the instruments to trade with. But when then applying the other systems, you could apply ADX there to get the best out of them.

My initial speculation here is not based on proper experience about ADX/ADXR. But this is boosting me to integrate these indicators into my small program and do some analysis together with VS. I'm starting to believe in a hypothesis that ADX and +/-DI might be a good filter for entering VS trades. I must also study it with respect to the exit and Dale's ideas about VSTOP. I would also like to see the impact of the used period length and whether the pairs behave differently with respect to it.

J.