Trend Following Research

[marketsurfer]

Quote from kut2k2:

<b>

We all "know" what trend is in a general sense; the key for at least some of us is finding a way to quantify it so it can be measured and exploited systematically. </b>

No question. That would be the holy Grail. Too bad it's impossible.

 

[kut2k2]

Quote from marketsurfer:

Quote from kut2k2:

We all "know" what trend is in a general sense; the key for at least some of us is finding a way to quantify it so it can be measured and exploited systematically.

No question. That would be the holy Grail. Too bad it's impossible.

That's what they said about supersonic flight. As to whether or not it would be the Holy Grail, that is debatable.

 

[Neenisti]

Quote from marketsurfer:

The post was directed at anyone who makes up definitions. Elite seems to be full of these folks.

Regarding Dan Rather--- I disagree with your analogy-- Dan rather interviewed people from all types of business, politics and every walk of life. Had he focused on a small niche such as trend following-- he would be an expert in that field. You are comparing a broad based journalist to a niche researcher. There is a huge difference.

Surf

There are plenty of individuals that focus their interviewing skills on the business field, the news media are full of them. None of them are considered experts on business.

Warren Buffett is considered an expert on business, he didn't get there interviewing business people.
Bill Gates is considered an expert on software, he didn't get there interviewing software people.
Stephen Hawking is considered and expert on theoretical physics, he didn't get there interviewing physicists.

 

[Trend Following]

Quote from Neenisti:

I would assume everyone understands the concept of buying pullbacks in extended up moves or selling pullbacks in extended down moves...

That's not trend following.

 

[Neenisti]

Quote from kut2k2:

Although I agree with Mr. Covel that knowing why price moves in a certain direction isn't necessary for a good trend trading strategy, you raise a good point in that a more explicit definition of trend would be enlightening in how best to exploit it. But I doubt we'll be getting any consensus on the definition. For example, the "long-term" reference in the wiki definition seems unnecessary because trend following can be done intraday, unless wiki is using "long-term" in a relative way, i.e., long within the given timeframe, no matter how fast that timeframe may be. For example, 200 minutes would be "long-term" for those trading in the 1-minute timeframe but wouldn't even register for EOD traders. Of course wiki gives no clue as to whether "long-term" is being used relatively or absolutely so the definition is subject to criticism.

We all "know" what trend is in a general sense; the key for at least some of us is finding a way to quantify it so it can be measured and exploited systematically.

I agree with the content of Mr. Covel's book too, as far as it goes. I think a more explicit definition can be found though.
My opinion is that the phrase, "long term" is relative to the chart as you state. This definition should be tested not just criticized.

 

[Neenisti]

Quote from Trend Following:

That's not trend following.

OK, so if that isn't a component trend following, and this marketsurfer guy considers you an expert on it, what is your definition of trend following? I'm not being argumentative just asking a question of an "expert".

 

[Gabfly1]

Quote from Trend Following:

That's not trend following.

Are you saying that there is only a very narrow definition of trend trading? If so, does that mean all trend traders essentially trade in the same manner, with little or no variance allowed for individual interpretation and exploitation of trend?

Since there seems to be some disagreement here on exactly what trend trading encompasses, and since you seem to be the resident expert on all things trend, then kindly define trend trading either as specifically or as generally as you think would be most useful for the purposes of this thread and discussion.

Edit: Neenisti, I see you beat me to it.

 

[Soon2Bgreat]

...and to think with All My Children going off the air I wouldn't get my daily fill of drama.

 

[marketsurfer]

Quote from marketsurfer:

Sure --- math proves that trend following does not work in the stock market or index futures. If it did-- confidence intervals, Serial correlation coefficients, regression coefficients of current changes versus past changes, and magnitudes of the impact of past moving averages on the future, distributions of the length of runs, the correllelogram, the expected waiting times between peaks and valleys, survival statistics could clearly show the edge. All these techniques are very good at discovering any non-random elements thus would "prove" the edge of trend following. They, in fact, indicate the opposite of positive edge when tested.

In addition, If you don't get basic stats --see " how markets really work" by Connors that uses simple studies to prove that an advance in a stock is far more likely after a series of lower lows than after higher highs-- the opposite of trend following.
Any questions?

The above lays out the commonly accepted basic statistical tests for trends ( non randomness). Unfortunately, trend trading fails the tests.

Surf

 

[Soon2Bgreat]

Quote from marketsurfer:

The above lays out the commonly accepted basic statistical tests for trends ( non randomness). Unfortunately, trend trading fails the tests.

Surf

Correllelogram?

Random-walks 'trend' according to most definitions (HH/LL etc.). Depends on what your definition of is is.