You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
What is the WORST indicator?
- Thread starter FishSauce
- Start date
Quote from inandlong:
That depends on the time frame. If you are looking to scalp off the 1 moment chart using stochastic crossover, then you are probably right. However if you are looking for a nice trend off the 5 and 15 minute charts, then I disagree.
Yes the time frame I was speaking of was seconds/scalping.
As for my opinion on 5 and 15 minute charts for daytrading...I just don't get it. By the time a 5 minute chart develops any sort of tradeable pattern, it's alreading 11:00am and the market is dying down into it's midday-chop-zone.
Of course it's sooo easy to look back at yesterday's action on a 5 min chart and pick out entry & exit points; but I'm talking about real-time, here and now.
I think stochastics is not all that bad when comparing two time frames. How do you all "read the tape" ? Where can us newbies learn to "read the tape" ? Can you actually spot good setups by reading the tape OR are you just scalping when reading the tape?
Let me give my view on this question.
Going back to another thread, a lot has been written and argued about "prediction" and "anticipation". In truth, we are dealing here with the same kind of problem. Nobody will probably find anything wrong with the proposition that both tape-readers and indicator users are doing this in order to "speculate" about a desirable outcome of their interaction with the market.
To bring some order in these ideas we should go back to basics. As some posts point out correctly, the tape is the primary data source containing transaction prices, volume. (One could extend this by allowing bid, ask and depth data). So the tape is mathematically the timeseries (vector timeseries if you wish).
The desire to "predict" time series is an old one. During the 2nd world-war Norbert Wiener developed what is known as the "Wiener Filter". Another major breakthrough was the "Kalman Filter". These filters allow the "prediction" in a probabilistic rigorous sense of "mathematically well-behaved" time series (and possibly some derived data).
The "mathematically optimal" solutions of the Wiener and Kalman filters are in fact equations that operate on the known data of the present to yield some "predictions" which can be proven to satisfy rigorous hypothesis.
The setup of the prediction of market data is not easily reduced to a "well-behaved" problem. In fact, what became known as the "moving average" was rather somewhat alike the optimum solution of Wiener filtering in some easy examples. Let me state again that no mathematical rigor could be applied to such market filter.
We are faced here with a very similar situation as the engineering applications of Wiener and Kalman filtering. Nobody disputes the fact that the primary data is the "tape". However, for the mission we want to accomplish, to speculate successfully or to shoot down an enemy plane, this data was not very useful to most people, speculators or gunners for that matter. The Wiener and Kalman techniques allow to "transform" or "filter" this tape datastream into one of more "indicators" more useful to accomplish the mission.
Our problem with indicators is that: (1) no rigorous modelling for markets seems to be available; (2) many people need to "predict" and/or "anticipate"; (3) "The Love Of The Marvellous And The Disbelief Of The True"which is the subject of another ET thread.
So TA, is simply the way people deal heuristically with "predicting" and/or "anticipating" the markets. One can argue endlessly about this without coming to any definite conclusion. Sometimes people can be very good at heuristics, for some tape may be ideal, for others "indicators"may help. Wiener and Kalman could indeed say what was "best" or "worst"in a mathematical sense. In this thread we cannot, let's all think about this.
nononsense
Going back to another thread, a lot has been written and argued about "prediction" and "anticipation". In truth, we are dealing here with the same kind of problem. Nobody will probably find anything wrong with the proposition that both tape-readers and indicator users are doing this in order to "speculate" about a desirable outcome of their interaction with the market.
To bring some order in these ideas we should go back to basics. As some posts point out correctly, the tape is the primary data source containing transaction prices, volume. (One could extend this by allowing bid, ask and depth data). So the tape is mathematically the timeseries (vector timeseries if you wish).
The desire to "predict" time series is an old one. During the 2nd world-war Norbert Wiener developed what is known as the "Wiener Filter". Another major breakthrough was the "Kalman Filter". These filters allow the "prediction" in a probabilistic rigorous sense of "mathematically well-behaved" time series (and possibly some derived data).
The "mathematically optimal" solutions of the Wiener and Kalman filters are in fact equations that operate on the known data of the present to yield some "predictions" which can be proven to satisfy rigorous hypothesis.
The setup of the prediction of market data is not easily reduced to a "well-behaved" problem. In fact, what became known as the "moving average" was rather somewhat alike the optimum solution of Wiener filtering in some easy examples. Let me state again that no mathematical rigor could be applied to such market filter.
We are faced here with a very similar situation as the engineering applications of Wiener and Kalman filtering. Nobody disputes the fact that the primary data is the "tape". However, for the mission we want to accomplish, to speculate successfully or to shoot down an enemy plane, this data was not very useful to most people, speculators or gunners for that matter. The Wiener and Kalman techniques allow to "transform" or "filter" this tape datastream into one of more "indicators" more useful to accomplish the mission.
Our problem with indicators is that: (1) no rigorous modelling for markets seems to be available; (2) many people need to "predict" and/or "anticipate"; (3) "The Love Of The Marvellous And The Disbelief Of The True"which is the subject of another ET thread.
So TA, is simply the way people deal heuristically with "predicting" and/or "anticipating" the markets. One can argue endlessly about this without coming to any definite conclusion. Sometimes people can be very good at heuristics, for some tape may be ideal, for others "indicators"may help. Wiener and Kalman could indeed say what was "best" or "worst"in a mathematical sense. In this thread we cannot, let's all think about this.
nononsense
I think stochastics is a dangerous indicator for newbies. Even George Lane emphasizes not to use it as a simple crossover type indicator. Also, it was originally designed for end of day grain data, where the relationship of closing prices to a lookback range could be significant. I question how valuable this is on one minute bars for stock indexes that are getting whipped around by programs.
Simple - your beliefs & emotions. Secondly, any one you don't really understand how to use effectively. Notice there is a distinction between knowing how to use an indicator and knowing how it works. The latter may be interesting, but the former is useful.Quote from FishSauce:
What is the worst indicator or sets of indicators?
I only use 2, but my guess is that in the right hands each indicator, and many undisclosed ones as well, serve a useful purpose.
My suggestion, however, is you don't get bogged down with to many. Find the one or combination a a very few that you can work effectively with and go stick to those exclusively.
I agree this is a strictly individual preference.Quote from nononsense:
... One can argue endlessly about this without coming to any definite conclusion. Sometimes people can be very good at heuristics, for some tape may be ideal, for others "indicators"may help...
I also agree that the tape contains the "real" information and that filters or indicators are merely used to help decipher the not so hidden story being told by the tape.
Now that I think about it, I never liked reading long stories, so maybe that's why my tape reading is confined to the past couple hours and my trades run typically less than 30 minutes.