Information diet

[Bad_Badness]

Does not really seem to be a question so I will just ramble on....

One thing to consider, since this is math, therefore logic based, is the concept of the contra posit.
Many are familiar with the concept: if A then B.

What is interesting about this is, as HUME, pointed out that A always proceeds B. I.e. empirically speaking Cause proceeds Effect. I.e B lags A.

The contraposit means that if there is no B, then there is no A. One can conclude if the effect is not there, then the cause did not occur also. Obvious right? (of course there could be if C then B, but that is different).

Then consider if an indicator is supposed to procced and "cause" a signal, then if the signal is not there, the indicator is not there either. With me still?

So it would follow that using a lagging indicator such as a MA that you want to give you a signal,
then consider the contraposit. If there is no signal, then the indicator is not there either. Seems pretty obvious, right!

So create an indicator that tells you "there is no signal", i.e. it tells you NOT to trade. Overlay that with your "regular" signals and see what you get. This method, will pretty much eliminate a lot of "indicators". What it eliminates is the "model" that your indicator is supposed to represent: If A then B.

A good model not only shows that if A then B but also if not B, then not A.

This is how I deal with the notion of "lagging".

Hope that helps and is not too nerdy! )

 

[destriero]

Willing to share what you are using?

Sorry, it was a joke, although I did sub to ransquawk or whatever in the early 00s on an unlimited cellular plan.

 

[2rosy]

Willing to share what you are using?

https://www.elitetrader.com/et/thre...re-is-a-new-entry-signal.290145/#post-4096049

 

[expiated]

So create an indicator that tells you "there is no signal", i.e. it tells you NOT to trade.

No question... Your comments simply bring experiences I've had with indicators to mind...

 

[schizo]

Now imagine going one step further and getting rid of charts, market timing, and market direction. Boom! Life changing when you begin thinking in terms of probabilities.

So you can think in terms of probability without looking at any charts, eh? You must be a genius.

 

[schizo]

I wanted a moving average that didn't lag, but found that those which were supposed to serve this function, like the Hull Moving Average and the Zero Lag Moving Average, were inadequate. I therefore set about learning enough about coding to design the indicators I envisioned for myself, if and when then did not exist. For example, this page from my book illustrates the difference between the two above-mentioned moving averages, and my own "instantaneous moving average."

View attachment 302466

Only way to create an instantaneous moving average is to set the parameter to "1". In that case, you will have the moving average move in synch with the close of the candle. Anything other than that, it will lag. Hence, I don't understand how you could have come up with "instantaneous" moving average (eg. when the price goes up, so does the MA). Care to elaborate further?

 

[Bad_Badness]

Interesting concept: instantaneous moving average. It could not be based on the close of a bar, by definition otherwise as Schizo points out it would be vacuous.

So one would have to base it on something before the close such as the average of the bar based on the VWAP, or the median, or mode. That would result is something that could be calculated within reasonable confidence level, before the close. Basically it has to somehow cheat time and be predictable before the close. But of course it would a moving mode or moving median or moving VWAP.

While possible, it would be better to choose a different concept than a Moving XYZ, or a different usage of the MA that does not have the problem of lagging. But it is hard to tell what one is trying to achieve in this hijacked thread.

Bottom line with Moving xyz, is often not that they are lagging, but how one uses them. IMO, it is better to accept the "lag", and work with it, not against it. To me it does a bunch of calculations that I don't want to do in my head. And I save the calculations that are too hard to code up, and are better done in my head. A properly trained mind is one of the most powerful analysis tools, one just needs to know how to use it, and avoid its inherent short falls. And make no mistake, it takes a lot of effort and formal training helps.

Hope this give you some ideas.

Indicators or not, is really not the issue. The issue is knowing which tools to use, how to use them, in combination for synergy. And know their weaknesses and pitfalls.

 

[Bad_Badness]

No question... Your comments simply bring experiences I've had with indicators to mind...

View attachment 302488

Nicely done chart!

 

[Sweet Bobby]

So you can think in terms of probability without looking at any charts, eh? You must be a genius.

Indeed I am. Charts are noise in my trading world.

 

[Laissez Faire]

So you can think in terms of probability without looking at any charts, eh? You must be a genius.

That’s generally how quants work. A chart is only a visualization of numbers. Those numbers and derived numbers/metrics can for example be put in a table allowing you to analyze/compare in a way that can’t be done with charts and with no need for charts either.