Using ADX as a filter to find trending markets

[pedro01]

Hi All

I apologise if this question is too much of a noob question for this forum but here goes anyway.

As background, I am developing my trading strategy which initially will be only dealing with trending markets.

I use Tradestation which as many will know has radar screen, so technical data can go on a list which helps to review markets en-masse.

When looking at a chart, it's quite simple to visually see which markets are trending or not but I was considering the merits of using ADX on the radar screen and looking for those in the 20-40 range as a method of shortlisting the markets I'd look into.

The obvious advantage is that in theory, I will have less charts to trawl through to find those in trends. Now - the drawback is that most recommendations I have had were to use an ADX setting of 14 against daily data, that's just 2 weeks, so on longer term trends you could just be seeing a retracement against the trend as a high ADX reading. It seems like a very short period to me.

Some ways I could tweak this is to use a longer period or a higher timeframe (say - weekly) but I am not really sure if this would devalue the ADX reading in some way.

Any thoughts on this ? I'm not using this as an entry signal, just to weed out trending markets.

Many thanks

Pete

 

[brutusdog]

Hi All

I apologise if this question is too much of a noob question for this forum but here goes anyway.

As background, I am developing my trading strategy which initially will be only dealing with trending markets.

I use Tradestation which as many will know has radar screen, so technical data can go on a list which helps to review markets en-masse.

When looking at a chart, it's quite simple to visually see which markets are trending or not but I was considering the merits of using ADX on the radar screen and looking for those in the 20-40 range as a method of shortlisting the markets I'd look into.

The obvious advantage is that in theory, I will have less charts to trawl through to find those in trends. Now - the drawback is that most recommendations I have had were to use an ADX setting of 14 against daily data, that's just 2 weeks, so on longer term trends you could just be seeing a retracement against the trend as a high ADX reading. It seems like a very short period to me.

Some ways I could tweak this is to use a longer period or a higher timeframe (say - weekly) but I am not really sure if this would devalue the ADX reading in some way.

Any thoughts on this ? I'm not using this as an entry signal, just to weed out trending markets.

Many thanks

Pete

hi pete,

i have found adx to be of little use as it lags substantially and therefor you end up in moves to late, if you want to understand trends and the constrution i suggest you read iraj's threads about cycles and cycle analisis as this is what trends are all about.

whatever way you choose to trade you want to be ahead of the game and have a statistical advantage over time preferably using leading or adaptive analisis or idicators much of which is included here.

 

[Grey1]

Hi All

I apologise if this question is too much of a noob question for this forum but here goes anyway.

As background, I am developing my trading strategy which initially will be only dealing with trending markets.

I use Tradestation which as many will know has radar screen, so technical data can go on a list which helps to review markets en-masse.

When looking at a chart, it's quite simple to visually see which markets are trending or not but I was considering the merits of using ADX on the radar screen and looking for those in the 20-40 range as a method of shortlisting the markets I'd look into.

The obvious advantage is that in theory, I will have less charts to trawl through to find those in trends. Now - the drawback is that most recommendations I have had were to use an ADX setting of 14 against daily data, that's just 2 weeks, so on longer term trends you could just be seeing a retracement against the trend as a high ADX reading. It seems like a very short period to me.

Some ways I could tweak this is to use a longer period or a higher timeframe (say - weekly) but I am not really sure if this would devalue the ADX reading in some way.

Any thoughts on this ? I'm not using this as an entry signal, just to weed out trending markets.

Many thanks

Pete

ADX is a lagging indicator which has the following disadvantages

1) poor entry hence you start your trade with large risk
2) Can only be used in Trending market which means

a) you don't know if the market is trending or not hence it gets you into many losing trades
b) Even if market did trend you still can make $$$$ 25% of the times as market only trends 25% of the times which means you MUST COUNT ON PERFECT EXIT to pay for your losses during the consolidating days which is 75% of the time.

3) ADX has zero predictive quality .. A cycle based oscillator however has some predictive qualities


Can I suggest for you to look into either CYCLE ANALYSIS or PRICE ACTION MOMENTUM EXHAUSTION theories to develop your trend following strategy

By Price action Momentum Exhaustion I mean

( SHORT STRATEGY )
IF Price breaks out of the range and continues in 3 WAVES of higher high each followed by consolidation AND a SHOOTING STAR or INVERTED HAMMER Formed at the last bar then SHORT

This is what is known in Advanced TA as Exhaustion and is often a very strong trend following reversal . This is how a new trend is BORN which LEADS the market and not like ADX which LAGS the market

IS this clear ? if not then it would be best to ask my in my seminar tomorrow as things like this must be demonstrated LIVE,, Far more effective




Grey1

 

[pedro01]

Thanks for the comments.

I don't plan to trade based on ADX, rather I was just considering using it on radar screen to filter out non-trending markets.

My trading plan involves identifying trending markets, confirming the trend is still valid & has strength & then getting in on retracements - I do not plan to attempt to take the whole trend, or a reversal although I will obviously use reversal signals as places to exit trades. I also plan to completely avoid choppy range bound markets which is why I will need to look across a number of markets as there will always be some markets with strong trends at any given time.

I will now study both CYCLE ANALYSIS & PRICE ACTION MOMENTUM EXHAUSTION and see how I can fit these in to my plan.

Thanks for the tips - I shall doubtless return once I have digested this material.

 

[pedro01]

After the seminar by Grey1 - I have officially abandoned my old plan....



I'm going through the archived ones now - although I can only find 3 of them archived.