MACD Divergence scan

[TSOKAKIS]

Quote from AAAintheBeltway:

skip it??? a divergence is the most powerful tool in tech analysis, if you know how to handle it. It has alwyas been difficult to scan for, as what constitutes a divergence is somewhat subjective. If the Amibroker scan can reliably pick it out I would say that is fantastic.

Here is a simple trading application :
11 stocks presented a positive [bullish] stochastic divergence on Jan 2, 2003.
See the results 6 bars later for this group in the attached gif.
Note that settings were
Buy, Sell at Open
Delay = +1
commission = 0.5%
Imagine the 60M shares who were selling CSCO on Jan 2, 2003and the 91M shares who were buying CSCO on Jan 10 !!
They have had just the opposite position, ignoring the Stochastic divergence and its application.
At least, one could avoid this mistake !!
Note that in the recent 3-year N100 history, a PDI indication >=7 is characterised as massive and should not be ignored for the whole market and, especially, for the stocks with the divergence.

 

[TSOKAKIS]

Quote from jack hershey:



skip it.

There are ALWAYS some profits behind a divergence. Will you skip them too ? Check again some exposed examples...

 

[TSOKAKIS]

Quote from jack hershey:



skip it.

A more than obvious Stochastic divergence for Light Crude Oil May contracts caused a drop from 36 to 30 in 5 trading days.
Is it something to ignore ??
Would it be better to buy CLK03 5 days ago ??
I do not think so...

 

[inandlong]

Quote from AAAintheBeltway:

skip it??? a divergence is the most powerful tool in tech analysis, if you know how to handle it.

Agreed.

 

[TSOKAKIS]

3 consequtive weekly MACD negative divergences on July99, Jan2000 and April2000 were real gold for ^GSPC traders.
A positive divergence on Oct2002 warned for an 8-week bullish period.
These are some examples of the importance of indicators divergences.

 

[TSOKAKIS]

A Mean indicator is the average of indicator values for the stocks of a Market.
Example:
MeanStochD=(StochD1+StochD2+...+StochDn)/n
You may see four examples and the detected bullish divergence date for the N100 market :
MeanCCI50, March12
MeanRelSlope, March12
MeanStochD, March13
MeanRSI, March12
The mutual confirmation was great at March12 close and anticipated the March13 bullish expansion.
Quick profits should be taken at yesterdays close, we are in a quite volatile period and there is no excuse for any delay.
3 of the 4 MeanIndicators bented at close, The MeanStochD will confirm today.
Note also that any other N100 MeanIndicator had a similar behavior.
It was not coincidence, it was one of the most clear signals ever met.
For your detailed reference you may see my MeanIndicators formulas at
http://www.amibroker.com/library/list.php

 

[TSOKAKIS]

Quote from TSOKAKIS:

A Mean indicator is the average of indicator values for the stocks of a Market.
Example:
MeanStochD=(StochD1+StochD2+...+StochDn)/n
You may see four examples and the detected bullish divergence date for the N100 market :
MeanCCI50, March12
MeanRelSlope, March12
MeanStochD, March13
MeanRSI, March12
The mutual confirmation was great at March12 close and anticipated the March13 bullish expansion.
Quick profits should be taken at yesterdays close, we are in a quite volatile period and there is no excuse for any delay.
3 of the 4 MeanIndicators bented at close, The MeanStochD will confirm today.
Note also that any other N100 MeanIndicator had a similar behavior.
It was not coincidence, it was one of the most clear signals ever met.
For your detailed reference you may see my MeanIndicators formulas at
http://www.amibroker.com/library/list.php

The next trading days after this post was a clear verification of the N100 behavior. This is the power of the MeanIndicators.

 

[jack hershey]

Quote from inandlong:


Agreed.

As you can see in one week of posting, no one gave the person asking a for a scan (using specifically TC2000) for divergence of MACD and price.

The div for an indicator (one with two lines at that) and a price presumably in a trend, is not a very good use of anyone's effort.

Many people spoke also about how difficult it is to use this concept (div between any indicator and price) in any application. It appears to me the requestor is pursuing new things and enlarging his scope of knowledge.

Skip it. I felt was a good viewpoint. Now it is a week later and the originator has not let us know how all the comments helped him out or if the technical crew at TC2000 gave him a fix.

 

[TSOKAKIS]

Quote from jack hershey:



As you can see in one week of posting, no one gave the person asking a for a scan (using specifically TC2000) for divergence of MACD and price.

The div for an indicator (one with two lines at that) and a price presumably in a trend, is not a very good use of anyone's effort.

Many people spoke also about how difficult it is to use this concept (div between any indicator and price) in any application. It appears to me the requestor is pursuing new things and enlarging his scope of knowledge.

Skip it. I felt was a good viewpoint. Now it is a week later and the originator has not let us know how all the comments helped him out or if the technical crew at TC2000 gave him a fix.

As you see from postings here, full divergence codes are available, work fine and give, IMO, interesting applications. For the special code of MACD div, take a look at

http://groups.yahoo.com/group/amibroker/message/36195
and att. gif
Unfortunately I do not know TC2000 language, but it is not an unsolved problem, the functions I use are widely available from many T/A softwares. A TC2000 expert could easily translate my divergence codes.

 

[jack hershey]

Quote from TSOKAKIS:



As you see from postings here, full divergence codes are available, work fine and give, IMO, interesting applications. For the special code of MACD div, take a look at

http://groups.yahoo.com/group/amibroker/message/36195
and att. gif
Unfortunately I do not know TC2000 language, but it is not an unsolved problem, the functions I use are widely available from many T/A softwares. A TC2000 expert could easily translate my divergence codes.

Sounds like you are focussed on MACD divergence. Neat topic.

The originator is focussed on MACD histogram (which is MACD Divergence and one other thing. The other thing is .....price.

That means this. He needs to scan to find a minimum value of this difference between price and the histogram value. Thank God it is not an angle.

I is my belief that a lot of responders did not address his need and desire. You think I am incorrect. I don't think that what he wants is worth pursuing, so I said so right at the beginning.

If you feel it is worth pursuing, you can see that a beginning still has to be made. He needs to plug the histogram and the price data into something and then select what is the necessary resultant value that, when he scans, triggers an entry on a list for his latr usage.