Jump to content

Welcome to the new Traders Laboratory! Please bear with us as we finish the migration over the next few days. If you find any issues, want to leave feedback, get in touch with us, or offer suggestions please post to the Support forum here.

  • Welcome Guests

    Welcome. You are currently viewing the forum as a guest which does not give you access to all the great features at Traders Laboratory such as interacting with members, access to all forums, downloading attachments, and eligibility to win free giveaways. Registration is fast, simple and absolutely free. Create a FREE Traders Laboratory account here.

jasont

Multiple Timeframes

Recommended Posts

Let me see?

(1) Market is not self-similar.

(2) Market is self-affinitive across different timeframes

(3) Market is pseudo-fractal.

 

 

Are we getting anywhere ?

Share this post


Link to post
Share on other sites
Let me see?

(1) Market is not self-similar.

(2) Market is self-affinitive across different timeframes

(3) Market is pseudo-fractal.

 

 

Are we getting anywhere ?

 

I've had bad experiences with anything starting with "pseudo-" ;) Besides what does that even mean "pseudo-fractal"...

 

Since you are a moderator, perhaps you can decide with jason what to do with the thread, since I'm pretty sure it's not anywhere near what jason expected of it right now (and I'm partly to blame for that as well). Change the title of this one and let him start over again elsewhere?

Share this post


Link to post
Share on other sites
I've had bad experiences with anything starting with "pseudo-" ;) Besides what does that even mean "pseudo-fractal"...

 

 

If you can't come up with a better one, then we are stuck with it. :rofl:

Share this post


Link to post
Share on other sites
jasont, maybe you should start a new thread called "Multi-Trend Strategies" and leave this one to discussing definitions and other mathematical concepts dealing with such a topic. Be sure to link this thread up front so as to make clear the difference between the two. Just a thought. ;)
Because we are currently in this situation (imo)... :hijacked:...maybe we should do the above. It may also get some new people involved who don't want to get into the current mix of splitting hairs. Even if we continue on now, someone is bound to come in and start things up again. Better to have a place where we can say "see here and reply there" to quickly stop it.

Share this post


Link to post
Share on other sites

Since you are a moderator, perhaps you can decide with jason what to do with the thread, since I'm pretty sure it's not anywhere near what jason expected of it right now (and I'm partly to blame for that as well). Change the title of this one and let him start over again elsewhere?

 

I think we can keep this thread as the way it was intended.

So far it was a good intro, in my opinion. But further theoretical discussion of Factal, Chaos, Mandelbrot.. etc will be considered off-topic. then a new thread should be created for that purpose. Agreed?

Share this post


Link to post
Share on other sites
I think we can keep this thread as the way it was intended.

So far it was a good intro, in my opinion. But further theoretical discussion of Factal, Chaos, Mandelbrot.. etc will be considered off-topic. then a new thread should be created for that purpose. Agreed?

So is it official, no more talk about the details of fractals? So for this thread, everyone just needs to agree to disagree. Hopefully we can all agree that the market is built off of multiple trends that are different sizes and these alignments and interactions are what we are here to discuss. :)

Share this post


Link to post
Share on other sites

Oh O the forum police have arrived.:rofl:

 

The practical applications of using multiple time frames (can we say MTF's?) are (imo) extremely powerful. Lets take a hypothetical trader that is attempting to catch intra day swings that might last under an hour to all day depending on market conditions. He's attempting to trade from (or close to) daily high to daily low (and back again) if the opportunity arises.

 

If he squashes up his chart or maybe calls up a higher time frame chart (perhaps hourly daily or weekly) he is in essence looking at a higher time frame. (When taken in reference to what he is trading i.e daily swings. He is looking farther out than his trading focus is if you like) This gives him a bigger picture at the expense of some detail. Many people do this either consciously or not to get some context for there trades. Areas of potential support and resistance are likely to be more significant if they have been tested and proven (or even broken) over the course of days or even weeks or months. They are likely too be 'stronger' than levels that have formed over minutes or even seconds. Looking at the bigger picture can give 'better' areas to trade against.

 

Conversely if price is in an area where our trader is interested in taking a trade and he calls up a faster chart than the one he watches routinely. Then he is looking at a lower time frame. He has given up some of the picture to examine the detail. The main reason to do this is to 'trigger the trade' or to try and read whether the area he is interested in is holding or breaking in (or closer) to real time. There is a potential danger, there is a tendency to remain 'zoomed in' for too long so he effectively ends up trading smaller swings, his focus has changed. I think this is a common malaise.

 

Of course there is nothing new here though it is probably worth mentioning Charlie Drummond who has arguably put forward the most rigorous and complete framework for multi time frame analysis. Still going strong in his 70's. Sadly his work is not in the public domain though some of the ideas have slipped out.

 

There got through the whole post without mentioning fractals....oh sh*t it just slipped out. :)

Share this post


Link to post
Share on other sites

Each time frame should have a function for the trader. Otherwise the trader will drive himself crazy with info overload and second guessing.

 

I use 3.

 

The highest is for general direction and to gauge what "type" of trade - breakout or pullback - depending on trend vs. range.

 

The middle is for reference as to where we are in the larger trend. Here is where I look for patterns that confirm the higher time frame.

 

the lowest is for precise trade entry - and where stops would go - in order to keep risk low.

 

hope that is on topic, and helps someone..

 

ws

Share this post


Link to post
Share on other sites
......Of course there is nothing new here though it is probably worth mentioning Charlie Drummond who has arguably put forward the most rigorous and complete framework for multi time frame analysis. Still going strong in his 70's. Sadly his work is not in the public domain though some of the ideas have slipped out....

 

The late Robert Krause did some good work in this area too. While I am not big on indicators, I like his use of one chart with indicators from higher timeframes. Mainly because it allows the trader to look only at the "entry, exit and stop" level chart-the small timeframe, while getting trend/momentum/support and resistance from the higher timeframes.

Share this post


Link to post
Share on other sites
Each time frame should have a function for the trader. Otherwise the trader will drive himself crazy with info overload and second guessing.

 

I use 3.

 

 

You are that hypothetical trader!!:) 3 is a good number having said that occasionally zooming out even further to get the even bigger picture can be useful. One way to use that particular information is to look for confluence.

Share this post


Link to post
Share on other sites

What has intrigued me earlier was Hlm's statement regarding how to go about measuring and estimating the current cycle of each time frame by using the concept of volume/market profile. The only things that appear to be cyclical to me in MP is the rotation around the VA and the magnitude of the rotation would be approximately measured by VAH-VAL. The next higher cycle may be from the first VA to a second VA, the price then may rotate back to the first VA or it may not.

Share this post


Link to post
Share on other sites
Each time frame should have a function for the trader. Otherwise the trader will drive himself crazy with info overload and second guessing.

 

I use 3.

 

 

The easiest way is to use the Heiken-Ashi indicators on 3 time frames:

 

(1) GREEN, GREEN, GREEN

(2) RED, GREEN, GREEN

(3) GREEN, RED, GREEN

(4) RED, GREEN, RED

(5) GREEN, RED, RED

(6) RED, RED, RED

 

6 :beer: That is how many days that took God to create the world.

Share this post


Link to post
Share on other sites
Oh O the forum police have arrived.:rofl:

 

There got through the whole post without mentioning fractals....oh sh*t it just slipped out. :)

 

haha, keep the fractal discussion going...I'm late to this thread and thought it was quite interesting, especially the stable levy distribution stuff.

I mean its hardly off topic to this thread to talk about MTF and fractals, its not like we are talking about multi time frames and a woman's ass...

One thing that I'm surprised has not come up is that this all depends on your style of trading. For my style, I'm not a trend trader at all and don't find MTF to be of any use. I don't like trend trading, I actually fear trends because by their nature you are giving up good position/price to get with the trend. The only time I ever trade with the trend is if a flag sets up. I view my style as a "level" trader...Its easy to find levels on a chart that have held importance to the market at the aggregate, and then I gamble as to the degree new information has been discounted or over priced as far as when these prior levels are tested by price again.

The danger of multi time frame analysis is in assuming a fractal nature to price and using the higher time frame pattern to make a prediction about a lower time frame pattern's path...To my mind that is worse than wrong, its absurd. While its easy to cherry pick a pattern that worked in this fashion in retrospect, what it ignores is that the information coming down the pipe at the present + 1 unit of time is not only unknown, its unknowable.

Probly an even more important point is to me this kind of discussion is where our tools kind of suck...we really shouldn't be having this discussion....Either a pattern at a higher time frame is self similar at a lower time frame and can make a prediction better than random or it can not. My guess is it can not but of course I may be totally wrong..if we had better pattern recognition software and all had monte carlo simulators we would be disussing the results of that experiment as opposed to guessing what the results of that experiment would be.

If you view it from that perspective, what is even the point of further discussion on this matter without those tools?

Share this post


Link to post
Share on other sites

It is worth clicking through to 'Lévy stable distributions' if you have an appetite for a bit of maths. Basically financial data series are fractal (by Mandlebrots definition) unless you disagree with his conclusions about the distribution of financial data series.

 

I don't know if its because I'm interested in this type of stuff but don't have the the math to really understand the math behind it that I've never ran across anything conceptually in this field that doesn't have a degree of absurdity to it.

What strikes me about modelling financial time series is your essentially modelling an X sided dice with each side having a different weight and then trying to model the distribution of the rolls of that dice..The problem though is both X and the weighting are moving in continuous time, along with the number of rolls, the force of how hard or soft you throw the dice...

While obviously you can get a Phd in statistics by modelling historic static dice rolls of this nature, it really has nothing to with the reality because your not modelling the important variables of the function, exactly because they are unknowable...so you fudge it to make your model work on paper.

Modelling the distribution is the easy stuff, the important stuff of modelling the variables is basically impossible unless there is enough computing power to find nash equilibrium in continuous time of an N player game, N equal to the number of players of the game, which obviously when it comes to the markets even knowing the number of players is impossible let alone the computation of something that can't be known.

That to me is quant finance in a nutshell.

Share this post


Link to post
Share on other sites
One thing that I'm surprised has not come up is that this all depends on your style of trading.
In my opinion multi-trend trading (the way I trade it) combines all three styles of trading depending on how you define them. You are trading with the trend, entering on a reversal of the smaller trend with your risk decreased, and you are holding a portion of your inventory for a breakout through the first smaller trend resistance to complete the cycle.

 

For my style, I'm not a trend trader at all and don't find MTF to be of any use. I don't like trend trading, I actually fear trends because by their nature you are giving up good position/price to get with the trend. The only time I ever trade with the trend is if a flag sets up.
It would be interesting to see a chart labeled with what you are looking at. Of course MTF is not for everyone. However, the majority of people I have explained it to resulted in a light bulb going off and their trading finally taking off. But again, more than one way to trade. :)

 

The danger of multi time frame analysis is in assuming a fractal nature to price and using the higher time frame pattern to make a prediction about a lower time frame pattern's path...To my mind that is worse than wrong, its absurd. While its easy to cherry pick a pattern that worked in this fashion in retrospect, what it ignores is that the information coming down the pipe at the present + 1 unit of time is not only unknown, its unknowable.
The up leg of a trend (see pictures I posted previously) is created by smaller up trends. The up leg of those smaller trends are created by smaller up trends. It's the nature of the market. It's the ebb and flow of the market. No, it's not perfect (of course), but that's what money management is for. I will not get into the discussion of self-similar because it's not needed. As for the predicting comment, there is no reason to "predict" the lower time frame. I think people are getting confused and stuck on this concept. There is no prediction, it's a fact...larger trends are created by smaller trends. The theory behind multi-trend trading is to find areas where you can enter with bias/pressure and limit your risk w/out limiting your reward. There is nothing to cherry pick if you are sticking to the true essence of a trend. If you are talking about the fractal concept shown on the pic that atto posted HERE, that is not what we are talking about.

Share this post


Link to post
Share on other sites
The late Robert Krause did some good work in this area too. While I am not big on indicators, I like his use of one chart with indicators from higher timeframes. Mainly because it allows the trader to look only at the "entry, exit and stop" level chart-the small timeframe, while getting trend/momentum/support and resistance from the higher timeframes.

 

Kraus was one of Drummonds students just fyi :) You can see the root of a lot of his work. Of course that doesn't diminish its value.

Share this post


Link to post
Share on other sites

Hi Jason

 

Nice post reg. multiple timeframes. Im trading in a similar way. Not using that many timeframes but otherwise the same. My idea is based on phases. 1-2-3-4 where 1 is the base before a run up. The run up is phase 2. Phase 3 is the base before the drop. The drop being phase 4. My idea is to trade the phase 1 and 3 breakout/breakdown on a lower timeframe but with the trend or range on the higher timeframe. If you are interested in sharing ideas let me know.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.