Moving Average Period - Preferences?

[Muschu]

Hi
I use a variety of MA periods when looking at basic charts.
What I have been tending to do is use a simple 2-line MA200 if looking at a 1-5 year chart; but both a 30 and 200 MA if looking at 1-2 month chart.
A question however: If I am looking at a SP over a relatively short period, say 1-2 months, is there a "preferable" MA setting? ie: Which is associated with greater significance? eg: As to when the MA is crossed by the SP?
Thanks
Rick

 

[Trembling Hand]

is there a "preferable" MA setting? ie:

YES. What ever you can make sense of. As in all trading setups that is all that should matter to you.

 

[Muschu]

YES. What ever you can make sense of. As in all trading setups that is all that should matter to you.

What makes sense to me may not make sense. I have no difficulty acknowledging that. Hence the question. Does anyone else have a view please?
Thanks
Rick

 

[wildkactus]

rick,
MA's I use are 5 / 23 / 233.
I use the 5 / 23 on daily and weekly charts, just to indicate graphically a change in trend. the 233 is there just to look at on monthly charts.

How I got these was by just changing the numbers until I got ones that made sense to me and worked with what I was looking for. You just have to find what is right for you, I don't think there is any right or wrong periods for an MA.
But remember they are a lagging indicator and I find that they are a good quick glance indicator.

 

[Muschu]

rick,
MA's I use are 5 / 23 / 233.
I use the 5 / 23 on daily and weekly charts, just to indicate graphically a change in trend. the 233 is there just to look at on monthly charts.

How I got these was by just changing the numbers until I got ones that made sense to me and worked with what I was looking for. You just have to find what is right for you, I don't think there is any right or wrong periods for an MA.
But remember they are a lagging indicator and I find that they are a good quick glance indicator.

Many thanks WK. Interesting to see what others do. I have been mainly using 30 and 200 and can get quite a different perspective as to where the SP sits in relation to the MA of course.

I am reading a book by Stan Weinstein - only 40 pages to date. He makes constant reference to a 200 [30 week] MA as an aid in decision-making. Haven't seen him use a shorter time frame yet.

I know they are a lagging indicator and serve a valuable purpose.

Thanks for the comment. I appreciate it.

 

[RobinHood]

Depends on your time-frame.


When I'm position trading US stocks:

I use 50 and 200dma over price, and 50dma over volume, for the reason that many institutions/players will be watching these and even use them as buying support.

 

[Whiskers]

Personally I didn't use MA's much until more recently when I came up with a combination of three different types that seem to be giving me good short and medium indicators to trade from. But I still don't use any MA longer than 14 days.

I rarely look at 90, 200 day MA's. I don't find them near as relevant in indicating probable future direction as MACD, Stochastic etc on a weekly or monthly chart.

I prefer Standard Deviation Chanels instead of 90 and 200's for medium term trends as a range guide in conjunction with FA for medium (month/s) to longer term (year/s) trends.

But having said that my main interest is in the more speculative and volatile end of the market. It may change if I get to a point when I decide to hold longer for dividends and take more advantage of the capital gain discount.

 

[tcoates]

Unfortunately, there are no magic numbers. You might not agree (?) with everything said in the video, but have a look at

http://www.youtube.com/watch?v=Z0Me9jkmWLg

I only use a 30 week moving average to get an idea of where it has been.

Tim

 

[howardbandy]

Greetings all --

You can runs tests to find out what works for the issues you plan to trade. Here is a short AmiBroker code segment that tests all values for the two moving averages, and buys when the first moving average crosses up through the second moving average. (It also computes the equity and plots the price, the moving average, and the equity. You only need the first six lines of code to test the system -- the rest is to give a visual plot of the results.)

Note that there is no restriction in the code to keep the first moving average shorter than the second.

If the "best" set of lengths has the first moving average length shorter than the second, then the results are telling you that you should be buying strength -- buy as the shorter moving average crosses up through the longer one -- the way most of the literature suggests. This is results in a "trend following" system.

If the "best" set of lengths has the first longer then the second, then the results are telling you that you should be selling strength -- buy as the shorter moving average falls down through the longer one. (Buy as the longer moving average rises up through the shorter one.) Opposite to common thought. This results in a "mean reversion" system.

So --- try it out for yourself.

In most cases -- for most tickers, most time frames, and most measures of goodness (the objective function) -- the "best" set of lengths will suggest that a mean reversion system works better than a trend following system.

One implication is that the common wisdom of buying when the price closes above its, say, 50 day moving average, has it backwards -- that is the time to sell.


//----------------------------
//
// MACrossSystem.afl
//
// The classic Moving Average crossover
// using simple Moving Averages
//

// Set up the lengths for the moving averages
Length1 = Param("Length1",6,1,81,2);
Length2 = Param("Length2",35,2,200,2);

// The Moving Average calculations
MA1 = MA(C,Length1);
MA2 = MA(C,Length2);

// The Buy and Sell logic
// Buy when MA1 crosses from below MA2 to above MA2.
Buy = Cross(MA1,MA2);
Sell = Cross(MA2,MA1);

// Compute the equity for the single ticker
e = Equity();
Maxe = LastValue(Highest(e));
Plot( Close, "Price", colorBlack, styleCandle );

// Plot the MA lines.
Plot(MA1,"MA1",colorGreen,styleLine);
Plot(MA2,"MA2",colorBlue,styleLine);

// Plot the Buy and Sell arrows.
shape = Buy * shapeUpArrow + Sell * shapeDownArrow;
PlotShapes(shape, IIf(Buy,colorGreen,colorRed), 0, IIf(Buy,Low,High));

// Plot the equity curve
Plot(e,"Equity",colorBlue,styleLine|styleOwnScale,0,Maxe);
Plot(10000,"",colorBlue,styleLine|styleOwnScale,0,Maxe);
GraphXSpace = 5;
//Figure 9.1 MA Cross System
//
//----------------------------------------------


Thanks for listening,
Howard

 

[Whiskers]

In most cases -- for most tickers, most time frames, and most measures of goodness (the objective function) -- the "best" set of lengths will suggest that a mean reversion system works better than a trend following system.

One implication is that the common wisdom of buying when the price closes above its, say, 50 day moving average, has it backwards -- that is the time to sell.

Thanks for that post howard.

I'm sometimes considered unorthodox, a bit of a black duck, but that sums up the problem I had reconciling MA's and conventional wisdom. I also had the overwhelming inner belief that one missed the boat getting on, and too often triying to jump ship on an outgoing tide by conventional wisdom.

I don't feel so 'odd' now, especially since my recently discovered kit of indicators tends more to your comments.

I'll be studying your rational and formula more closely now.

 

[tech/a]

Personally I ask what is it your trying to achieve with an M/A?

Answer that then select your M/A.
Frankly if you cant read what your stocks doing without an M/A--what are you doing?
That said they have their place as a point of reference.

 

[Muschu]

Personally I ask what is it your trying to achieve with an M/A?

Answer that then select your M/A.
Frankly if you cant read what your stocks doing without an M/A--what are you doing?
That said they have their place as a point of reference.

The point I was trying to ascertain is just that -- their place as a point of reference. There seem to be many proponents of the fact that MAs do have a place and I was trying to make some further determinations in that direction. Are there not some who consider MAs a critical aspect of analysis as there are others who may not? Perhaps there are even some who would argue that they can read the performance of their stocks without reference to any chart whatsoever.
May he/she who has the perfect system look down upon us mere mortals from on high.

 

[tcoates]

Rick,

Here are some numbers for you...

* Elders (in his second book, whatever that was called) looks at 26 week moving average.

* Articles on the ASX web site suggest using a 30 week moving average

* Weinstein uses (/used) a 30 week moving average.

* Jim Berg uses a 34 week moving average.

Think these would all be EMA. And it depends on how you want to use an moving average.... I use it as a reference point. That's because I find it faster to look at the moving average than to look at the price bars.

To me, it is about looking at the bigger picture. If you can access myasx newsletters then have a look at...

http://www.asx.com.au/resources/newsletters/investor_update/20071016_moving_average_secrets.htm

Whether you plot a weekly moving average over a daily chart, or just look at a weekly chart with the appropriate moving average(s) on it.

FYI - using the above moving averages there is a difference on 75c between the highest and lowest average for BHP. I don't tend to look at value of the moving average, just the direction.

Tim

PS. Good on Tech for not needing a moving average. Some of us just find it easier looking at a chart with one present.

PPS. Tech - it is the how you use a moving average on a chart that makes the difference between traders. Just because some like a moving average does not make then better or worse than you seem to indicate in your response re needing a moving average.

 

[wavepicker]

Personally I ask what is it your trying to achieve with an M/A?

Answer that then select your M/A.
Frankly if you cant read what your stocks doing without an M/A--what are you doing?
That said they have their place as a point of reference.

Excellent point tech, IMO like chasing shadows moving across a wall!!

 

[Muschu]

Many thanks Tim for your constructive response which actually attempted to answer the question.

It seems that many people of acclaim and experience find a MA of value - in one form, another, or several.

If others, experts or otherwise, find a MA useless, vague, immaterial, boring, innaccurate, off-putting, illusionary, delusionary, inferior, shadowy, or outside the set of parameters they choose to use in their strategies... then so be it. Each to their own. I'm yet to learn of a perfect system.

Tim your references were very helpful; your comments valuable and to the point. Thank you.
Regards
Rick

 

[tech/a]

There are 4 lines in my reply.
You have chosen to focus on one.

I'm yet to learn of a perfect system.

Perhaps your search maybe in an inappropriate area (and I dont mean specifically M/A's)

 

[Sean K]

Perhaps this thread should be in the Trading Strategies or Beginners Forum...

Whatever the case, lets keep it friendly, and not be too critical of each other for either not having an advanced understanding of technical indicators, or otherwise.

 

[Muschu]

There are 4 lines in my reply.
You have chosen to focus on one.
Perhaps your search maybe in an inappropriate area (and I dont mean specifically M/A's)

Tech I apologise if I focused on only one line - I did not mean to do so.

By using an MA I am trying to add to my learner's pool of information.
I can read certain things about stocks without an MA but find the MA useful.
And, yes, I use an MA as a reference.

As a learner I comfortably acknowledge that I have far more questions than answers. Explanations that people offer are very welcome.

Regards

Rick

 

[Rockhoundnz]

I use 5 and 20 period MAs on my 3 min, 15min, and 60 min charts, and an 89 period MA on my 89 tick chart.

 

[wayneL]

I use 5 and 20 period MAs on my 3 min, 15min, and 60 min charts, and an 89 period MA on my 89 tick chart.

What about on all the time-frames in between?

utthedoor: