Help with ADX/DMI

February 16th, 2015, 03:26 PM

thank you again for your response.
I have been basically looking at the ADX/DMI combo in order to get rid of some of the whipsaws that happen when trading with the PSAR / MACD.

I will defenitly look into the others you have mentioned in your reply.

About the time frame I intend to trade we are on the same page it seems. Long swing/position trading is what I am interested in mostly.

What about the TTM_Squeeze? Do you see any advantages on using it as a filter to avoid ranging/whipsawing periods?

yes, I am a photographer. Mostly travel and street photography. It is very relaxing indeed to just go out and look at the world from that perspective after days of looking at charts. These two activities seem to complement each other somehow. I know a lot of traders that love photography as well. interesting...

Underexposed

I stick mostly to daily charts...I see the benefits of weekly charts in certain situations but I find that a 6 month daily chart is all I really need. For longer term I use a P&F chart which I use to view macro resistance/support and can see several years in one chart depending on the volatility of the share price. It is important to know the absolute high's for a stock as these are powerful resistance points as traders have elephant memories for stuff like that.

I do use intra-day charts fairly often.... but not for daytrading (I find that too stressful for my liking). What I use is generally a 5 or something 2 day candlestick chart with an interval of either 5min or 15 min. I use such charts to plan an entry or exit with respect to a position.... or when I want to establish a Stop level.

There I can see mini resistance/support levels and I can try to get a stock slightly cheaper or squeeze a few more pennies from a sale.

It takes a lot of practice to learn interpretation of charts properly. the approach I use right now is the result of an evolution of over 20 years of study on my part.... That is why you see odd-ball parameters such as 30 day lookbacks. I have tested and retested these indicator combinations and charts. A lot of people think one indicator will solve all problems.... it won't but a consensus of a suite of not-the-same-look indicators works for me.

It takes time and study... every situation with a stock is different... you must assemble the clues... like solving a mystery puzzle.

You might think that having done all this work I would stand pat... but I am still open to new approaches to see if they would fit in with what I already do.

BTW...are you a photographer??? I do a fair bit of nature photography to relax.... birds, butterflies and wild flowers

February 16th, 2015, 04:55 PM

ositokillao

About the time frame I intend to trade we are on the same page it seems. Long swing/position trading is what I am interested in mostly.

then really learn how to use my "trigger Chart"...that is really the only one needed for a swing trade... from a few weeks to a few months

ositokillao

What about the TTM_Squeeze? Do you see any advantages on using it as a filter to avoid ranging/whipsawing periods?

no... I don't find any advantage to Keltner's Channels... if you use Bollies (and your should really learn them and the associated indicator BBwidth) this is all you need.

I don't use ThinkOrSwim ... I don't use any platform at all really... I have my charts that available through several free charting areas and I use a discount broker for my trades. I took a look at their Learning center and I definitely do not agree with how they use Bollinger bands or BBwidth.

I am in this forum because I really like talking about indicators and how I use them.

I think I have directed you to the spot in my Canadian journal where I discuss the construction of my "trigger chart" and how it is constructed...Keltner's channels would just clutter the chart for no advantage to me.

Others may find them useful but I don't... I can find most everything they do using other methods.

February 17th, 2015, 09:45 PM

Here's a video of Hubert Senters he shows how to use the ADX/DMI in a different way.

How to Use Power Shift Indicator - Hubert Senters

To plot this indicator on the Thinkorswim platform just use the DMI indicator study and in the property box change the DI+ and DI- to draw as Histogram instead of a line and to plot the 20 line just go into edit source and insert this code plot Line20 = 20; Now you'll have the powershift indicator like Hubert Senters shows.

February 17th, 2015, 11:34 PM

Thanks LukhyG for the "heads up" on Hubert Senters' PowerShift indicator. I put together the TOS thinkscript below for any/all whom might be interested...

Code

declare lower;

input length = 14;
input averageType = AverageType.WILDERS;

def hiDiff = high - high[1];
def loDiff = low[1] - low;

def plusDM = if hiDiff > loDiff and hiDiff > 0 then hiDiff else 0;
def minusDM =  if loDiff > hiDiff and loDiff > 0 then loDiff else 0;

def ATR = MovingAverage(averageType, TrueRange(high, close, low), length);

plot "DI+" = 100 * MovingAverage(averageType, plusDM, length) / ATR;
"DI+".SetDefaultColor(color.DARK_GREEN);
"DI+".SetLineWeight(3);
"DI+".SetPaintingStrategy(PaintingStrategy.HISTOGRAM);

plot "DI-" = 100 * MovingAverage(averageType, minusDM, length) / ATR;
"DI-".SetDefaultColor(color.RED);
"DI-".SetLineWeight(3);
"DI-".SetPaintingStrategy(PaintingStrategy.HISTOGRAM);

def DX = if ("DI+" + "DI-" > 0) then 100 * AbsValue("DI+" - "DI-") / ("DI+" + "DI-") else 0;

plot ADX = MovingAverage(averageType, DX, length);
ADX.SetDefaultColor(GetColor(4));
ADX.SetLineWeight(3);

plot "20Line" = 20;
"20Line".SetDefaultColor(GetColor(9));
"20Line".SetLineWeight(1);

Hope this proves useful/helpful...

February 18th, 2015, 05:23 PM

thank you. I am starting to see things a bit clearer now with BB and BBwidth.

Underexposed

then really learn how to use my "trigger Chart"...that is really the only one needed for a swing trade... from a few weeks to a few months

no... I don't find any advantage to Keltner's Channels... if you use Bollies (and your should really learn them and the associated indicator BBwidth) this is all you need.

I don't use ThinkOrSwim ... I don't use any platform at all really... I have my charts that available through several free charting areas and I use a discount broker for my trades. I took a look at their Learning center and I definitely do not agree with how they use Bollinger bands or BBwidth.

I am in this forum because I really like talking about indicators and how I use them.

I think I have directed you to the spot in my Canadian journal where I discuss the construction of my "trigger chart" and how it is constructed...Keltner's channels would just clutter the chart for no advantage to me.

Others may find them useful but I don't... I can find most everything they do using other methods.

February 18th, 2015, 05:39 PM

netarchitech

Thanks LukhyG for the "heads up" on Hubert Senters' PowerShift indicator. I put together the TOS thinkscript below for any/all whom might be interested...

Code

declare lower;

input length = 14;
input averageType = AverageType.WILDERS;

def hiDiff = high - high[1];
def loDiff = low[1] - low;

def plusDM = if hiDiff > loDiff and hiDiff > 0 then hiDiff else 0;
def minusDM =  if loDiff > hiDiff and loDiff > 0 then loDiff else 0;

def ATR = MovingAverage(averageType, TrueRange(high, close, low), length);

plot "DI+" = 100 * MovingAverage(averageType, plusDM, length) / ATR;
"DI+".SetDefaultColor(color.DARK_GREEN);
"DI+".SetLineWeight(3);
"DI+".SetPaintingStrategy(PaintingStrategy.HISTOGRAM);

plot "DI-" = 100 * MovingAverage(averageType, minusDM, length) / ATR;
"DI-".SetDefaultColor(color.RED);
"DI-".SetLineWeight(3);
"DI-".SetPaintingStrategy(PaintingStrategy.HISTOGRAM);

def DX = if ("DI+" + "DI-" > 0) then 100 * AbsValue("DI+" - "DI-") / ("DI+" + "DI-") else 0;

plot ADX = MovingAverage(averageType, DX, length);
ADX.SetDefaultColor(GetColor(4));
ADX.SetLineWeight(3);

plot "20Line" = 20;
"20Line".SetDefaultColor(GetColor(9));
"20Line".SetLineWeight(1);

Hope this proves useful/helpful...

Very useful, thanks. It is basically a DMI/ADX using histograms.

However, if we are to wait for the ADX to be above 20 in order to enter a trade, a lot of big moves are lost. If the DMI crossover occurs but the ADX is still under 20 that doesnt always mean that there is no trend. The ADX lags after the crossover.

There must be a better way to confirm the trade so that it is not caught too late.

February 18th, 2015, 09:24 PM

ositokillao

thank you. I am starting to see things a bit clearer now with BB and BBwidth.

Thank you very much for your vote for my instruction

As far as ADX goes I will reiterate my thought on ADX when I said

Quoting

In truth I don't put a lot of faith in the ADX value itself... it is the DI+/- values that are important to me. Specifically the values where the look back is 30 days... not 8 days or 14 days

I do use the DMI (also known as DI) a lot....they are very reliable and most informative.

But ADX which is a strength of trend but not the direction of a price... Also it also lags the action by quite a bit in digging deeper into its calculation I see the reason.

This chart comes from Stockcharts.com showing the calculation of DI(14) and ADX(14)

It takes 14 days of data to calculate the first DI+/-....it takes 14 more days to calculate the first ADX(14).

No wonder there is a lag between the ADX and DMI... the DI+/-(14) values are not related to the ADX at all since that ADX is a simple average of the previous DX values calculated from from previous DI+/- values.

It is no wonder that I did not find the ADX useful...

In an intra-day chart where you are focused on what is happening RFN (right now) you are basing your conclusions on pretty heavily smoothed data.

February 21st, 2015, 10:53 AM

Underexposed

Thank you very much for your vote for my instruction

As far as ADX goes I will reiterate my thought on ADX when I said

I do use the DMI (also known as DI) a lot....they are very reliable and most informative.

But ADX which is a strength of trend but not the direction of a price... Also it also lags the action by quite a bit in digging deeper into its calculation I see the reason.

This chart comes from Stockcharts.com showing the calculation of DI(14) and ADX(14)

It takes 14 days of data to calculate the first DI+/-....it takes 14 more days to calculate the first ADX(14).

No wonder there is a lag between the ADX and DMI... the DI+/-(14) values are not related to the ADX at all since that ADX is a simple average of the previous DX values calculated from from previous DI+/- values.

It is no wonder that I did not find the ADX useful...

In an intra-day chart where you are focused on what is happening RFN (right now) you are basing your conclusions on pretty heavily smoothed data.

thank you.

I see now why ADX lags and how its true importance is to point out trend strength but taking this in consideration...

I was wondering if there is a way we can detect ADX slope angle? in degrees I don�t think is possible but how about in percentage from the previous bar?

I would love to be able to scan for ADX value AND angle of slope. I have played around with the scans but there is nothing like it.

Say for example a SCAN that would give me all stocks with ADX of 20 and sloping up by 25%

Is this possible? thanks.

Thanks.

February 21st, 2015, 11:12 AM

I am referring to this article here:

Direction is the Key to Using ADX Correctly

I think they mean ADX actual value when referring to .25, .50 and 1.00 not percentage...

so how about a scan that will detect an increase in ADX ( selectable ) ?

You could ask it to scan for ADX increasing a full point from the previous bar for example....

also the scan should be able to set the starting point for the matching stocks...

say...

scan for stocks with an ADX of 18 or higher and with a current adx value that is one point higher than the previous bar...

Sounds simple enough.. anybody care to code this???

March 1st, 2015, 12:11 PM

Christhegoods

this is a really good idea. have you seen aliens extraterrestrial visual systems thread at ff? he shows an awesome use of adx on MTF. i'd love to simplify adx by using the angles like you suggest on mtfs. ultimately i'd love to make it into a currency meter. coders should be all over this idea. imo, adx is one of if not the most reliable indicator there is. the reason it is not more popular is that it is very hard to read. i have trouble reading it. i hope someone who can code weighs in here.

hi, no, I havent seen the thread.

Can you please, send me the link?

I couldnt find it using the search.

I am a graphic designer by trade and I do see some advantages as to have more elaborate design clues for indicators that can actually pinpoint setups better and earlier.

I am very interested in seeing that thread.

Help with ADX/DMI

Discussion in ThinkOrSwim

Help with ADX/DMI