Moving Average

[motorway]

While Price must Fluctuate

Price does not have to trend (up or down )

To have a trend you need two units of fluctuation in the same direction.

eg HH & TT

Prices trend because of waves of buyingandselling ( this is the one thing there is no selling wave or buying wave there is only a buyingandselling wave )

We talk about buying and selling as if they were separate things that are not connected....

because trends occur because of buyingandselling
There is a movement up and down alternating

The direction of these ups and downs is the trend

Waves have peaks and troughs...

A pullback is a small trend
of again at least two fluctuations
That does not retrace the last trend movement by 100% or more
( tech's trend strength )
and does not lead to a congestion pattern ( can be expanding or dampening )


Prices cycle between trend and range ( a series of 1 unit fluctuations that alternate HTHTHTHT or THTHTHTH)

The best predictor of a trend is the absence of one
+ active ranging ( nothing to do with moving average or so many bars etc )

Why ? because the fluctuations in prices have/create, memory --> fear / hope

Trends are fractal , on one scale what is a 1 unit fluctuation might be a 4 unit fluctuation

Hence prices that are ranging on one scale
tend to be trending on another particular scale

Define your unit of fluctuation ( unit of risk and reward )

Identify strength and weakness ( Like what is the background --> season )
( eg retrace and duration etc )
Identify ranging or trend ( OK ranging can also be called a sideways trend )

nothing to do with "time frame" or moving averages of "timeframes"

Two units of fluctuations = a trend
alternating single units = range

A trend on one scale is a range on another..
scale is an invariant ,always optimized ( will not work with timeframes---> invariant ,always non-optimized )

The way up must be the same as the way down


The building of expectation ( ranging )
the repletion of expectation (trending up or down )

craving and satiety




motorway

 

[fapturbo]

Waves have peaks and troughs...

Define Peaks and Troughs so that it can be quantified

 

[kam75]

I have been looking at heaps of charts lately and finding the 200 day MA to have some merit. A lot of stocks have broken the 200 MA and continued upwards. Some bounce off it and some well haven't reached it and others see it as resistance

What got me into MA as an indicator was the reading of Sam Weinstein's book. He drills it home on nearly every page.

He recommends 30 Wk MA for investors and 10 Wk MA for traders.

Just recently the XAO pulled above the 200 MA which some say is a bullish sign.

Anybody prefer one MA tmeframe over another?

Hi Nero64

I'm a big fan of Weinstein's book and have been using his 30 week MA as a primary filter in my trading system to ensure stocks I buy are entering or already are in stage 2 (ie, above a rising 30 week MA). It's a brilliant and simple way to ensure you put your money into stocks that have a potential to rise. Stan's the man.

regards

 

[beamstas]

Define Peaks and Troughs so that it can be quantified

You ask alot of questions..?
Tell us the answer

 

[bunyip]

Unless there's a blackout in Sydney and IB goes down, while the SPI stays up!

My stop is put in place automatically when I enter the trade. Blackouts don't affect me.

 

[bunyip]

Hi Nero64

I'm a big fan of Weinstein's book and have been using his 30 week MA as a primary filter in my trading system to ensure stocks I buy are entering or already are in stage 2 (ie, above a rising 30 week MA). It's a brilliant and simple way to ensure you put your money into stocks that have a potential to rise. Stan's the man.

regards

I agree.
Not only does Weinstein's methodology give you a timely entry into stocks that are rising and are likely to rise further, it also takes you out before your profits can turn into brutal losses.
And it shows you how to profit on the short side when the trend turns bearish.

There have been many posts on this forum over the last 18 months from people who suffered catastrophic losses from the market slump.
Those big losses could have been avoided if they'd been familiar with Weinstein's strategy, and had implemented it.

Pity Manny Cassamatis from Storm Financial didn't read Weinstein.

 

[bunyip]

why does everyone fight this so hard?

price can be predicted - not 100% right, 100% of the time.
but to a higher degree than random chance.

I cannot see how anyone is making money if they cant predict future prices.

if you couldnt predict price... why would u ever enter or exit a position?

It's an interesting debate.....when you buy, are you predicting higher prices or aren't you? And can your prediction achieve an acceptable degree of accuracy?
One school of thought is that you would only buy into an existing trend if your prediction was higher prices.
Another school of thought is that someone taking a position in an existing trend is predicting nothing, he's simply reacting to what prices are doing right now, without knowing or making any predictions about what they'll do in future. He simply takes a position in the hope that the current price action will continue. He has the potential to make big profits if it does continue, or the prospect of a small loss if it doesn't.

I no longer enter into that debate. I simply say that you stack the odds in your favour by going with the highest probability. A new trend is likely to continue further in the same direction. It doesn't matter whether you call that a prediction or a probability or anything else.
You get a huge edge when you identify the dominant trend and then throw yourself in front of it.

 

[motorway]

It's an interesting debate.....when you buy, are you predicting higher prices or aren't you? And can your prediction achieve an acceptable degree of accuracy?
One school of thought is that you would only buy into an existing trend if your prediction was higher prices.
Another school of thought is that someone taking a position in an existing trend is predicting nothing, he's simply reacting to what prices are doing right now, without knowing or making any predictions about what they'll do in future. He simply takes a position in the hope that the current price action will continue. He has the potential to make big profits if it does continue, or the prospect of a small loss if it doesn't.

I no longer enter into that debate. I simply say that you stack the odds in your favour by going with the highest probability. A new trend is likely to continue further in the same direction. It doesn't matter whether you call that a prediction or a probability or anything else.
You get a huge edge when you identify the dominant trend and then throw yourself in front of it.

Every Entry is a anticipation or it is blind gambling

In your example why not enter long in a down trend ?

We can not predict Outcomes
But we can anticipate Conditions

That is why you go long in the (early) uptrend
and why you plant Tomatoes at the end of Winter and not Summer.

You can anticipate and identify conditions

conditions are cyclical

winter summer / Bear market Bull market


The outcomes are not predictable
So manage risk
and steer through to the outcome


------

If we are using an invariant like "unit of fluctuation" that means something
instead a variant like bars on a bar chart

All sorts of relationships can be objectively determined
by quantity,,

1 unit up 1 unit down = range
2 units up 1 unit down = trend
4units up 3 down = still trend but weaker

1 unit up 5 down = collapse into downtrend

peaks and troughs have an actual existence that have a singular location in Time and Space....

because you are using objective criteria

You can define nothing real with x number of bars
you can not objectively define anything
because when you change the time scale
The way up is no longer the same as the way down

relationships change--> variant

When you condense a 4 unit up and a 2 unit down move
into a 2 unit up and a 1 unit down ( double the unit of risk and reward )

You have The same pattern in the same location
in the same CONTEXT..

Invariant ---> completely real

If the 4 unit up took 2 weeks
and the 2 down took 1 day

when you move to a monthly chart
the pattern and its context and relationships is no longer visible

You would need to chart every up and down move in its own time frame
and then condense those muti time frames by the same %

Then you would have an objectively invariant chart
where peaks and tops have existence
and do not alter when the frame is changed

The fluctuations on the tape have definite location in time and space
that means everything can be objectively defined as long as the way up is the same as the way down..

motorway

 

[kam75]

Those big losses could have been avoided if they'd been familiar with Weinstein's strategy, and had implemented it.

Pity Manny Cassamatis from Storm Financial didn't read Weinstein.

Yeah the fund managers should get themselves a copy.

 

[Mr J]

Define Peaks and Troughs so that it can be quantified

They can be, but obviously it means using clear rules rather than subjective intepretation.

I no longer enter into that debate. I simply say that you stack the odds in your favour by going with the highest probability. A new trend is likely to continue further in the same direction. It doesn't matter whether you call that a prediction or a probability or anything else.

I'd liken it to inertia.

 

[fapturbo]

You ask alot of questions..?
Tell us the answer

A peak has to describe the highest point between two levels.

As a trough has to describe the lowest point between two levels.

Which levels??

How about a definition of a short term change in direction.

Short Term Up Trend and a Short Term Down Trend.

Then the Peak would be the highest point between the Change from short term up trend to short term down trend.

Likewise the Trough would be the lowest point between a Short Term Down Trend and a Short Term Up Trend.

Short Term Up Trend we could use the bars highs and lows and we could say that a short term up trend is when there is two or more consecutive bars that have higher highs.

Short Term Down Trend could be when there is at least two consecutive bars that have lows that are lower.

Once we have a change from one Short Term Trend to another then we can check to see what the highest/lowest price is between.

Then when we can find the highest price this can be the Peak. The lowest price can be the Trough.

Yes or No ???

 

[beamstas]

A peak has to describe the highest point between two levels.

As a trough has to describe the lowest point between two levels.

Which levels??

How about a definition of a short term change in direction.

Short Term Up Trend and a Short Term Down Trend.

Then the Peak would be the highest point between the Change from short term up trend to short term down trend.

Likewise the Trough would be the lowest point between a Short Term Down Trend and a Short Term Up Trend.

Short Term Up Trend we could use the bars highs and lows and we could say that a short term up trend is when there is two or more consecutive bars that have higher highs.

Short Term Down Trend could be when there is at least two consecutive bars that have lows that are lower.

Once we have a change from one Short Term Trend to another then we can check to see what the highest/lowest price is between.

Then when we can find the highest price this can be the Peak. The lowest price can be the Trough.

Yes or No ???

I know what a peak and what a trough is, the question you asked is how to quantify it?

 

[fapturbo]

I know what a peak and what a trough is, the question you asked is how to quantify it?

Thought I just did

Code so that check for what i said??

 

[Mr J]

A real life example is less clear fapturbo. The line between a peak, trough and noise varies for everyone. The same chart will be interpreted differently by different traders. Then we must also consider the context, as what we identify a peak in one chart we may lablel as noise on another.

 

[tech/a]

You cant use peak and trough code for system testing as each is dynamic.
You dont know a peak or a trough is in place until well after they are formed.
Zigzag is the most common indicator used to formulate peak and trough.

 

[fapturbo]

You cant use peak and trough code for system testing as each is dynamic.
You dont know a peak or a trough is in place until well after they are formed.
Zigzag is the most common indicator used to formulate peak and trough.

A trough is formed as soon as there is two consecutive bars which have higher highs. Assuming there was a Short Term Down Trend in place previous to this. That is assuming there was at least two bars prior to this that had lower lows.

The trough does not move once it is set.

What is a trend? Isn't it a series of higher Peaks and higher Troughs??

We could use the troughs and peaks as levels on the chart to trade from?

For example if price goes below or closes below a previous trough for example we trigger a short??

 

[tech/a]

I understan what you are saying so these 2 charts show a "setup"
One short and one long.
If trading is above the peak shown then a long is triggered
and trading below the low trough then a short is triggered.

The "setups" are on the FAR right of the chart.

This is how you see it?

 

[fapturbo]

Something like this...

Once we have a trend in place we look for set up like this to trade.

 

[fapturbo]

I understan what you are saying so these 2 charts show a "setup"
One short and one long.
If trading is above the peak shown then a long is triggered
and trading below the low trough then a short is triggered.

The "setups" are on the FAR right of the chart.

This is how you see it?

Yea that would be it.

 

[fapturbo]

Chart of XJO

Next Peak is not known yet as there is not a pattern of two consecutive lower lows.