Measuring Trend

[HATEtheRisk]

Quote from schizo:

I know what a TREND is when I see it. I just don't know until I see it.

agree

 

[Joman]

Quote from schizo:

I know what a TREND is when I see it. I just don't know until I see it.

Very true. Why bother to measure a trend ?

Whatever the method you will use, you will measure something of the past and the risk the trend is about to end is large.

 

[kut2k2]

Quote from Joman:

Why bother to measure a trend ?

Enter on a strong trend, exit on a weak trend. How did you think trend trading worked?

Quote from Joman:

Whatever the method you will use, you will measure something of the past and the risk the trend is about to end is large.

So you never take a trade unless you're guaranteed a win? Okey dokey.

The risk of the trend ending is large only if you're late measuring the trend. This is why my own research has focused on optimal noise reduction with minimal lag. Not as "impossible" as it sounds if decent adaptive smoothers are utilized.

 

[Joman]

Quote from kut2k2:

The risk of the trend ending is large only if you're late measuring the trend. This is why my own research has focused on optimal noise reduction with minimal lag. Not as "impossible" as it sounds if decent adaptive smoothers are utilized.

Noise reduction with minimal lag sounds great , I want to learn more about it

As far as I'm concern I think you will always be late measuring a trend. If you're able to measure it, it means it already has happened.

Consider the market as an alternance of consolidations and trends: you would prefer knowing / measuring when you have a consolidation pattern and prepare for the next trend rather than measuring the current trend.

 

[kut2k2]

Quote from Joman:

Noise reduction with minimal lag sounds great , I want to learn more about it

Google is your friend

Quote from Joman:

As far as I'm concern I think you will always be late measuring a trend. If you're able to measure it, it means it already has happened.

Then what good would noise reduction with minimal lag do for you?

Quote from Joman:

Consider the market as an alternance of consolidations and trends: you would prefer knowing / measuring when you have a consolidation pattern and prepare for the next trend rather than measuring the current trend.

Why doesn't your same logic apply to consolidations as it does to trends? "By the time you can measure it, it's already over and you're too late."

 

[baro-san]

Quote from kut2k2:

Enter on a strong trend, exit on a weak trend. How did you think trend trading worked?
So you never take a trade unless you're guaranteed a win? Okey dokey.

The risk of the trend ending is large only if you're late measuring the trend. This is why my own research has focused on optimal noise reduction with minimal lag. Not as "impossible" as it sounds if decent adaptive smoothers are utilized.

Or: reverse when week trend, and hold when strong trend.

Noise is when you don't know what is happening. It's yours. So, smoothers are crutches that prevent you from getting better.

 

[baro-san]

Quote from Joman:

... As far as I'm concern I think you will always be late measuring a trend. If you're able to measure it, it means it already has happened.

Consider the market as an alternance of consolidations and trends: you would prefer knowing / measuring when you have a consolidation pattern and prepare for the next trend rather than measuring the current trend.

If you watch carefully, consolidation is a weak trend. Then when a strong trend weakens, it signals its end.

 

[kut2k2]

Quote from baro-san:

Or: reverse when week trend, and hold when strong trend.

You're trying to pick tops and bottoms. Good luck with that.

Quote from baro-san:

Noise is when you don't know what is happening.

I know exactly what's happening: the current trend has ended and the next one is unpredictable, which is why I'd never try to pick a top or a bottom.

Quote from baro-san:

It's yours.

No, it's the market's. I'm not big enough to create any noise in the market.

 

[Stoxtrader]

Quote from kut2k2:

Analysts often run to LR because it's easy, not because it's truly appropriate or even "good enough" lots of time.

In measuring financial trend, you have three options:

Option #1 - Account for the curvature of data somehow.

Option #2 - Make irrelevant the curvature of the data.

Option #3 - Just use a linear regression and hope for the best.

The third option is the worst by far. The KER uses the second option and works better than LR as a result.

KER?

 

[kut2k2]

Quote from Stoxtrader:

KER?

Kaufman Efficiency Ratio