Moving Averages - Dr Shane Oliver

[Muschu]

Many thanks Timmy and MRC:
Timmy: Thanks for going to so much trouble in your explanation. It made excellent sense and was exceptionally well put. Rational, logical, not emotive, knowledgeable and educational. Best I've read for someone in my position.
Your writing, despite your comment, is in fact very clear. [Alternative / additional career path maybe?]
Back at the being of this thread I made some reference to longer term traders. That was partly because, as a SF retiree, I am not into day trading. [Hence my interest in the 200MA]. I probably should have made this much clearer. Sorry. There's a section in your post where you mentioned that "many opt for FA and a longer holding period". This is my position although I am trying to incorporate a basic knowledge of TA as well.
Yes, I noted Wayne's success - although I think he later modestly referred to that particular period as a "purple patch". Good on him.
However I have learned a great deal from your comments and really do appreciate the time and trouble you put into making them.
Your closing paragraph is intriguing and I'd be interested to learn more of your views at some future time. When I look at promotions for share-trading based seminars [that awful term "wealth creation"] or newsletters I am inclined to wonder if the some of these promoters in fact earn more from these activities than they do from the market itself. [Please don't assume I am putting them all in the same boat.] And if a broker confidently publicises 10 strong buys then why live off the brokerage -- why not keep quit, invest and live off the profits? Much easier. [I'm aware that there will frequently be other issues on the agenda]. However perhaps they are just complementing their income in case their trading goes awry. Doesn't bother me - I just have an interest in behaviour and, sometimes, in the assumptions we make about others and their motives.
I learned a huge amount from your comments T - thanks mate.

MRC: Many thanks for the references. I've noticed many positive comments about Nick Radge on ASF and have briefly skimmed some of his own posts. He appears to come very highly recommended in all regards. If his seminars were suitable for someone in my [retiree] position then I'd be interested in learning more about them -- if he comes to Perth that is.
I recently got a copy of Stan Weinstein's book [Profiting in a Bull or Bear Market - or words to that effect]. Once I manage to digest that I'll look into the books you mentioned. I have noted the titles for future reference. Many thanks.

Perhaps I'll let Shane Oliver RIP now - his comments at least served a purpose in that I am now better informed.

All the best

R

 

[wabbit]

The use of Technical Analysis (TA) versus Fundamental Analysis (FA) has always caused discussion (arguments), so you'd think that all the TAers would be like-minded, and the FAers to be like-minded too? Not so.

Even in the TA ranks there are those who believe in Gantt, Elliot and other "hocus pocus"; some believe in Fibonacci, some like Wilders etc. Some people don't believe in the random walk theory, some do. Some TAers like to smooth out the "noise" to discover the underlying signals; some TAers trade the noise by detrending the data. No-one will ever agree on everything; it just causes arguments. This is especially so when it comes to trading.

Allegedly, moving averages are one of the most widely used TA tools. (I forget where I read this, so cannot reference it, but it sounds pretty much right from my observations of the past 11 years of trading) This might be because they are "simple concepts" that people can understand (people don't like maths for some reason?), it could be other reasons. I think the main reason that people use these techniques is to smooth out the "jaggedness" of a chart. There have been many psychological studies into the different ways that traders view charts, much of this research has shown that as humans, traders like to see things as simply as possible. A smooth curve is much more pleasing to the eye than a jagged graph.

John Ehlers and others get into some "heavy" maths when describing finite impulse response filters, infinite impulse response filters and other digital signal processing techniques which can (?) be applied to markets; they are referring to simple moving averages and exponential moving averages but using big words and lots of maths/engineering "speak" and provide some interesting details on their applications. Ehlers mentions MA lag as a tool that can be exploited by traders. He and others have shown methods to remove lag from MAs. (If this works, there should be the same chart of the XAO (daily) as motorway displayed on the previous page, but this time a "lagless" 200 period EMA is applied.)

The TA trader that uses MAs to filter out the noise might say the XAO has turned a corner and is now heading up, so take long positions. The detrended-filter TA trader might say the difference between the index value is diverging from the MA, so enter long. Ultimately both of these traders will take up long positions; why they chose to do so may be dependent on how they see the same 200 period MA, but neither of them will agree on what it means!

With MAs being in use (allegedly) so widely, there is the distinct possibility that these tools become self-fulfilling prophecies. I am not aware of any rules that say when the price approaches the 200 period simple moving average from above that everyone has to start buying because the price is approaching the 200 period moving average from above! Just because it happens, time and time again, doesn't mean it will happen again next time; next time the price might just keep descending and end way lower than the MA as if it weren't even there! TAers will have different opinions on what causes that too, and often blame it on the FAers!

Traders must be aware of the LIKELY behaviour of the markets in these regions where "majorities" have been noticed to lurk in the past. How you specifically get into a trade, or how you see the market may depend on moving averages, Elliot waves, Fibonacci retracements, PER, profit announcements, directors buying/selling, chicken bones, crystal balls, soothsayers, throwing darts at the finance pages etc. but careful risk management and trade management are the skills that any trader needs to be disciplined in if they are to survive.

0.02

wabbit

 

[motorway]

That was partly because, as a SF retiree, I am not into day trading. [Hence my interest in the 200MA]. I probably should have made this much clearer. Sorry. There's a section in your post where you mentioned that "many opt for FA and a longer holding period". This is my position although I am trying to incorporate a basic knowledge of TA as well.

If most things people do is lag

what is it they lag

Of all lines on a chart that can be drawn ( indicators included )

What is the one line that either projected or shifted

( projected = extended , Shifted = moving the entire line forward of backward )
Does not change its value ?

And even with this be careful how you use it.

Here is a free e-book
that a longer term investor working from fundamentals
and wondering about the why of Technical approaches
might find interesting .

http://www.clayallen.com/Winning The Performance Game 12-16-2006.pdf


motorway

 

[Kauri]

I have always found the 193.7034 MA
to be a good one...
until everyone else..
cottons on to it ...
that is...

Cheers
...........Kauri

:bier:

 

[dhukka]

In this morning's West Australian, Dr Shane Oliver, AMP chief economist, is quoted as saying:
"From a technical perspective both global and Australian shares are looking overbought after their strong rally from mid-March and they are now up against technical resistance associated with their trailing 200-day moving averages."

A question: Is a 200 day MA a period of time more likely to be emphasised and used by "longer term" traders?
[I'm assuming that most analysts use a variety of periods but am wondering whether shorter-term or day traders place their greater emphasis on a much shorter MA period].

Comments very wlecome and appreciated.

Thanks

R

You mean the same emminent Dr Oliver that predicted the All Ords would crash through the 7000 mark in 1Q08? Why people put so much credence in what the so-called experts say, who get it as consistently wrong as they do right, continues to amaze me.

 

[Muschu]

You mean the same emminent Dr Oliver that predicted the All Ords would crash through the 7000 mark in 1Q08? Why people put so much credence in what the so-called experts say, who get it as consistently wrong as they do right, continues to amaze me.

I'm sorry but this is really dreadfully silly. I had planned to be quiet on this thread but...
On ASF I have read posts from people saying that the All Ords will reach the high teens; or that they won't buy BHP until it is under $20 -- Are these perhaps self-appointed experts?
At times it strikes me that there is a real competition to be the gloomiest prophet out there; to be a better expert than the experts; while, at the same time, professing to being a highly successful trader -- best of both worlds with all bases covered --
I don't know if Dr Oliver would describe himself as "eminent" - I suspect not.
When did he make this prediction? What was the wording? Where is the evidence that he is wrong as often as right? [He may be, but please deomonstrate it.]
What differentiates a "so-called expert" from a real expert? What criteria defines the latter? Who establishes these criteria?
In this uncertain world isn't there a place for us all?
I am as amazed by your view - as you are by the views of "people". It certainly seems easier to criticise than to praise.

I'm almost sorry, apart from the positive comments received [which I hope I have duly acknowledged], to have started this thread with what I thought to be a simple question -- not a statement of commitment to any strategy or guru.

 

[Muschu]

If most things people do is lag

what is it they lag

Of all lines on a chart that can be drawn ( indicators included )

What is the one line that either projected or shifted

( projected = extended , Shifted = moving the entire line forward of backward )
Does not change its value ?

And even with this be careful how you use it.

Here is a free e-book
that a longer term investor working from fundamentals
and wondering about the why of Technical approaches
might find interesting .

http://www.clayallen.com/Winning The Performance Game 12-16-2006.pdf


motorway

Thank you Motorway. I enjoyed this.
I also saved the 243 page book to my favourites for reference. At first glimpse it looks very interesting.
Regards
Rick

PS - I thank everyone for your comments and will leave this thread be now.

 

[dhukka]

I'm sorry but this is really dreadfully silly. I had planned to be quiet on this thread but...
On ASF I have read posts from people saying that the All Ords will reach the high teens; or that they won't buy BHP until it is under $20 -- Are these perhaps self-appointed experts?
At times it strikes me that there is a real competition to be the gloomiest prophet out there; to be a better expert than the experts; while, at the same time, professing to being a highly successful trader -- best of both worlds with all bases covered --
I don't know if Dr Oliver would describe himself as "eminent" - I suspect not.
When did he make this prediction? What was the wording? Where is the evidence that he is wrong as often as right? [He may be, but please deomonstrate it.]
What differentiates a "so-called expert" from a real expert? What criteria defines the latter? Who establishes these criteria?
In this uncertain world isn't there a place for us all?
I am as amazed by your view - as you are by the views of "people". It certainly seems easier to criticise than to praise.

I'm almost sorry, apart from the positive comments received [which I hope I have duly acknowledged], to have started this thread with what I thought to be a simple question -- not a statement of commitment to any strategy or guru.

Now I'm amazed at your emotional fragility. The reference to 'eminent', is one that I made myself on this forum last year. Oliver is almost always referred to as 'Dr Oliver'. I believe the title conveys an air of expertise. I did a quick google on Oliver and one of the first references named him as an expert.

I find this amusing for a couple of reasons. As anyone in academia knows, (providing they have the capacity to be honest with themselves) all a doctorate really says is that you devoted three or more years of your life to some narrowly defined specialist area and produced something that will probably never be read by anyone nor is it readable to anyone except other so-called experts.

Secondly, as anyone familiar with the topic of economics knows, it is little more than a speculative area of investigation. That Nobel prizes are awarded in the field of economics to people whose theories are then blown apart only 10 years later attests to that. But I digress.

Dr Oliver made the call on the XAO cracking 7000 in 1Q08 back at the end of July. Here is the quote:

Market is just correcting itself

Mr Oliver says over the next six to 12 months the market will continue to rise, as the $US100 billion loss represents only 1 per cent of the total outstanding mortgages in America.

Despite yesterday's almost 3 per cent fall to 6082.9 points, Mr Oliver is forecasting that the benchmark ASX 200 index will reach 6700 points by year end and break 7000 points in the first quarter of next year.

"There will a rough patch for the next few months but the market will rally hard in the last few months of the year," he said.

Do a search on this forum and you will turn up other quotes from Oliver from last year that turned out to be horribly wrong.

Also check out the rather tragically named Oliver's Insights. They only have 6 months worth of history so fortunately for Oliver you can't see the magnitude of the errors he made last year although you can still probably find some gems.

When I got in to the stockbroking industry in 1997, Oliver had already established himself at AMP as chief economist and because AMP at that time was (and still is?) by far the biggest money manager in Australia, sell side analysts generally listened to what he had to say. I don't have a running scorecard of his track record but I can remember a fair amount of derision going his way.

The guy makes predictions on the short to medium term movement in the stockmarket. That by its very nature means that he is often going to be wrong. I have nothing against discussing Oliver's comments, however the opening post seems to take for granted that Oliver's opinion is worth something more than the average punter. I think that assumption should be viewed with a degree of skepticism given Oliver's recent (evidence provided) and historical (no evidence provided) track record.

 

[dhukka]

The above is not meant to imply that all academics produce nonsense of no value. Of course there are exceptions and some amazing stuff is written that provides great insight. However, that is very few and far between IMHO.

 

[Muschu]

Hi Dhukka

Can we perhaps terminate our discussion by agreeing upon the following:
- Dr Oliver's quote was pre-August/September. Like many others he failed to predict the future accurately. Fine.
- We have differing views on the value of most PhDs. Fine. [I hope my own post grad studies in psychology, education and business were not without purpose].
- Similarly we have differing views on the value of economics. Fine.
- Stockbroking may also contain an element of investigative speculation.
- Perhaps we are both easily amazed and amused. Don't know.
- Emotionally fragile? Big call based on little info. I don't mind.
- I acknowledge that I did not seach ASF to locate your earlier use of the term "eminent" in relation to Dr Oliver.
- Original post contains an element of assumption? Given that it consists of a newspaper quote and a simple question, I am satisifed to leave this comment of yours to others to make up their own minds.

Bottom line is that it seems we disagree and I am sure we can both live with that.

Regards

Rick

 

[tech/a]

AND

A moving average is still a line on a data set plotting a period average selected.
Reagardless of how you plot it.
Nothing more nothing less.

 

[howardbandy]

Greetings all --

One of the postings to this thread mentioned Dr John Ehlers. Here is a link to Ehlers' home page, which has links that will let you download his PowerPoint presentations and indicator code. It is considerably more complex than moving averages.

http://www.mesasoftware.com/index.htm

Thanks,
Howard

 

[howardbandy]

Greetings all --

We can test this. Here is the AmiBroker code.

// Test200dayMovingAverage.afl
//
// Is it profitable to Buy when the closing price rises
// through the 200 day moving average, and Sell when it
// drops down through the 200 day moving average?

MA200 = MA(C,200);

Buy = Cross(C,MA200);
Sell = Cross(MA200,C);

Short = Sell;
Cover = Buy;

Plot(C,"C",colorBlack,styleCandle);
Plot(MA200,"MA200",colorBlue,styleLine);

PlotShapes(shapeUpArrow*Buy + shapeDownArrow*Sell,
IIf(Buy,colorGreen,colorRed));


I ran this program using a list of the current stocks in the S&P 500 composite. The list I have has 497 entries. The date range was 1/1/2000 through 5/16/2008. End of day data, trading on the close of the signal bar.

For Long-only: 78 tickers showed a profit, 415 a loss. Most losses were completely bankrupt.

For Short-only: 9 showed a profit, 485 showed a loss.

So ---- should we Buy when the price rises through it's 200 day moving average? Probably not.

Thanks,
Howard

 

[Muschu]

So ---- should we Buy when the price rises through it's 200 day moving average? Probably not.

Fascinating Howard. Thanks.

Would it be valid to test this with say only 20 diverse and hand-picked stocks from the ASX100 with a 30 or 50 day MA? I have no idea - but could nominate the stocks if you think the exercise has any merit. If it has zero merit please just say so.

Thanks for the contribution.

Rick

 

[motorway]

http://www.mesasoftware.com/Seminars/TSWorld05.ppt

non linear , adaptive , median , phases , cycles



FRACTALS even...


But, it just looks like a complicated way of doing Point and Figure.

( superior or inferior is a valid question ... P&F is 100% all of the above without any optimization or lagging of parameters )

The PPT reminds me of a 1926 book .

P&F sssshhhh

motorway

 

[howardbandy]

Fascinating Howard. Thanks.

Would it be valid to test this with say only 20 diverse and hand-picked stocks from the ASX100 with a 30 or 50 day MA? I have no idea - but could nominate the stocks if you think the exercise has any merit. If it has zero merit please just say so.

Thanks for the contribution.

Rick

Hi Rick --

The "talking heads" regularly make statements like "yyy stock is above its xxx day moving average and that is bullish."

Using whatever trading system development platform you wish, vary the "200" in the 200 day moving average code in the system above and see if the stocks you are interested in are profitable in the days immediately following the bar where the price crosses up through the moving average. Or where a short term moving average crosses up through a longer term moving average.

Then try the other direction. Try buying when the price, or a short term moving average of the price, crosses down through a longer term moving average.

In my experience, most common stocks do not show a profit when a moving average crossover is used to initiate a trend-following trade. But, most do show a profit when a moving average crossover is used to initiate a mean-reversion trade. The holding period for the mean-reversion trade will usually be just a few days, or perhaps a week or two.

Be careful to avoid "optimizing the symbol space." If I try a system on 500 stocks and it works on 450 of them, I might have discovered something universal and potentially profitable. If it works on only 50 of them, then I have a system that does worse than random and I should not trade even those stocks where it showed a profit. (Unless there is something unique about those 50 so that they act in a similar fashion, but no others do.)

If I hand pick 20 stocks that show a profit, and they are diverse, then I am probably fooling myself. If I hand pick 20 stocks that have a common theme (like they are all gold producers), then they will tend to act the same. In fact, an index made up of those 20 will be much easier to model than any of the individuals. For example, look at the nine S&P Sector Exchange Traded Funds -- XLB, XLE, XLF, XL... -- it is fairly easy to develop models that trade these profitably.

Thanks,
Howard

 

[howardbandy]

But, it just looks like a complicated way of doing Point and Figure.

motorway

Hi Motorway --

When you have time, read one of Dr. Ehler's books or work through some of the examples from his presentations and technical papers, then code them for use in your favorite trading system platform. They are considerably more powerful than point and figure.

In my opinion, techniques such as he describes will play a big part in the future of mechanical trading systems.

Thanks,
Howard

 

[motorway]

Hello Howard,

have you done any work on P&F ?

Which is really a digital filter... discretization by price( % or arithmetic ) rather then by time.

tIME then become the speed of unfolding Time

and there is no need to switch between frames of tIME

eg We can represent a P&F chart as

a series of 1s & 0s

00011011101111001111

These binary digits represent price , but dynamic price

So there is speed acceleration and higher derivatives ( eg jerk )

and it is like a flow of a river.... So it dams up builds potentials etc

We can also see smoothness or turbulence

eg a very solid ( perserverating ) up pattern

is 011011011011011011011011011011011


P&F are two dimensional price charts with time as speed and not frame

eg We do not watch a wild animal every x time periods
because it changes rhythm---It changes speed and IT STRIKES..
So we better watch for units of movement and preparation...

To the extent I have looked at one dimensional price charts ( trying to make them non linear , adaptive , timely )..they are not as useful

you do not know ahead of time when the "strike" will occur ( so what time frames are relevant ) But we do know what the "striking distances" are and what are relevant movements.. If it moves we act in time.


hence I find moved based price discretization ...Two dimensional price charts
that incorporate time as is should be ( non linear )

in other words P&F very powerful

As much broken rhythm and speed change as there can be
P&F is on the job
because it is 100% adaptive there are no gaps

IF it moves there is something moving it..

motorway

 

[Muschu]

I wish some of you guys spoke English... This would be a great help to me.
Seriously, I appreciate your comments and observations but I'm starting from scratch over here.
Thanks
Rick..

 

[MRC & Co]

ha ha, your not alone Rick!

Many here use various methods, so when someone starts talking to you about their method, in the language of that method, it can become very confusing

The more you learn, the more you will understand, but there will always be many things posted here you still have no idea about!

It's really about finding the strategies you think appear the most realistic, you enjoy and then finding your niche.