Any indicator derived from price contains less information than price. This can be useful, if you want less information. But to be consequent, you would than have to drop price and trade without. Is that reasonable?
The idea I have, is not creating a panoply of different indicators from price, but to take new information - not the same - and use one indicator on that new information.
What do you think would be more useful for trading
- some lines identifying support and resistance
- volume or bid/ask-traded volume
- the NYSE Tick
- price of another index future (ES is the leader)
- the US-Dollar Index or the Euro
Trade off 1 parameter revisited and repackaged 5 times, or trade off 5 different parameters, each of them displayed once?
I believe in volume and intermarket analysis. The midline of the Donchian Channel for the NYSE tick is a better trendfilter than an exponential moving average. You get additional information on market breadth, which is not provided by price action for TF. The EMA cannot tell you anything more than your price chart, but less.
I get your point, but I think like any dogma the no-indicator movement can also be extreme. The EMA can tell me something about past price which I don't have the screen real estate to see without scrolling, and if I do scroll then I can't see the current price.
The following user says Thank You to fluxsmith for this post:
you are going to get instances where all your option (1) indicators are teling you to go long and your option (2) indicators are mixed and you are caught in the headlights. One will tell you long, other short. It all sums to nil.
Don't get me wrong, I'm no fan of RSI or Gann or whatever... but all of these techniques in option (1) and (2) and in volume alalysis and in intermarket analysis are using the same information.
My girlfriend says "baby, it's not what you got, it's what you do with it" - in technical analysis, we all have the same. It's what you do with it.
Like I said, "people" see only what they want to believe.
-> If I use a longer lookback period than the visible fraction of the chart, they can transfer information from prior periods. My fib confluence indicator uses a lookback period of 300 days and then transports that information on my 5 min chart. The pullback oscillator I use, has a period of 144 bars. The pivot indicators display information from the prior day, prior week or prior month. So these indicators add new historical information to the chart.
But the idea is to use fewer indicators. I am writing this, because I have to convince myself to drop some of them.
"the EMA cannot tell you more than the price itself"...
Sure, the EMA is a function of the prices that you have set it to. But your quote is like saying that volume bars don't tell you anything more than reading the tape will. Fact is you can't do the sums fast enough, same as an EMA gives you ballpark.
Here I do not agree. The NYSE Tick gets you additional imformation. If it shoots up, stocks have left their Gaussian distribution and show positive correlation. This can only be achieved
- by a news release
- by arbitrage
If there is no news, it means that the index futures have gone up too far. You may expect a snapback, if the NYSE tick exceeds 900 (actually I am normalizing this to adjust a bit to the daily trend as measured by the NYSE tick). So you get a valid countertrend signal. I have put this on audio, so if I hear the music I know that it is time to check for an exit, if I hold a long position. Depending on the mood of the day you could also enter short.
This information is not available from the price chart of the index futures. It is genuine new and additional information. But maybe you don't want to believe that, see above.
Well, OK you're right that NYSE TICK isn't the same as a futures price.
All of this information is from the same population - information produced by the market. This is the information that technical analysis uses, and it's all a function of time and sales across all market instruments.
The Gaussian distribution stuff... lol whatever floats your boat matey!!
Sorry, I did not want to disturb you with Gauss, don't want to be blamed for your nightmares. Tried to say, usually there is sort of random behaviour, some stocks being sold others being purchased. If they all move together, either it is the news of the day, or it is a buy programm triggered by HFT algorithms.
Technical Analysis feeds on feedback, creates new feedback, and feeds that feedback back to where the feedback came from.
The following user says Thank You to Fat Tails for this post: