Experiments in Technical Analysis

kkseal · #**921** 27th April 2007, 09:01 PM

Thanks Uashish & Saint. Actually, it's just a small involvement and doesn't deserve such praise. To be honest didn't begin with any great noble intention (like 'social committment') either. Just don't like selling my books to 'kabadis' and didn't quite know what to do with an old P1 machine that was taking up precious space in my small 2BHK apartment - so donated (thinking at least they'd be of some use to someone). Thereafter became friends with some of the volunteers (they are the real dedicated folks who deserve all the praise) and got a little more involved. That's all there is to it. But yes, even this little involvement is 'soul satisfying' particularly when i see the children. They are so happy with so little.

Regards,
Kalyan.

P.S. : I am sure Saint Karthik wouldn't mind this little digression.

karthikmarar · #**922** 27th April 2007, 11:09 PM

Quote:

Originally Posted by Saint

Hi Kalyan,

ps:Karthik,sorry for this post,a minor distraction in this great thread.....but I couldn't help it.

Keep up the great effort,and all the best,my friend.

Saint

Minor distraction..... It is a privilege to see you post in this thread, my friend. It is a pleasure to see you around…was wondering what happened to you ..since you stopped replying to my mails…

warm regards

Karthik

asnavale · #**923** 30th April 2007, 11:16 PM

CONTINUING WITH MACD

Hello friends,

As I was away on tour, I could not continue my posts in this thread about the MACD. I am now continuing the same now.

In my earlier post I had described the MA lag and methods of removing the lag. One way to remove the lag is taking two MAs with periods in the ratio 1 to 2 as described earlier. Another method is to take the MA for a period p and then taking the MA of this MA itself for the same period. The first calculation of MA introduces a lag of p/2. The second calculation (MA of MA) introduces additional lag of p/2. Therefore the difference between the MA and MA of MA results in a lag of p/2. So if we add this difference to the first MA we get the zero lag MA. This is the procedure used to calculate the zero lag MACD.

As MACD is difference between a short term EMA and a long term EMA (default periods being 12 and 26) the zero lag MACD is calculated by taking the zero lag EMAs. Similarly the zero lag signal is calculated by taking the zero lag EMA (default period is 9) of the zero lag MACD. The complete AmiBroker AFL code for calculating the zero lag MACD and the signal line are given below:

Code:

p1 = Param("Short period", 12, 3, 50, 1);
p2 = Param("Long period", 26, 5, 100, 1);
p3 = Param("Signal period", 9, 3, 25, 1);

EMAshort = EMA(C, p1);
EMA2 = EMA( EMAshort, p1);
Diff = EMAshort - EMA2;
ZLEMAshort = EMAshort + Diff;

EMAlong = EMA(C, p2);
EMA2 = EMA(EMAlong, p2);
Diff = EMAlong - EMA2;
ZLEMAlong = EMAlong + Diff;

ZLMACD = ZLEMAshort - ZLEMAlong;

EMA1 = EMA(ZLMACD, p3);
EMA2 = EMA(EMA1, p3);
Diff = EMA1 - EMA2;
ZLSignal = EMA1 + Diff;

Plot(ZLMACD, "zero lag MACD", colorGreen);
Plot(ZLSignal, "zero lag Signal", colorRed);

The result of this calculation is shown in attached Example-10 (Those who need the chart in only black & White may see attached BW_Example-10). For keeping the chart simple, I have not drawn the Histogram. In this figure the upper chart is the normal MACD and the lower one is zero lag MACD. The price chart is at the bottom. We can see that the peaks and troughs in Zero Lag MACD are sharper and occur earlier than in normal MACD. The MACD and Signal line cross-overs also occiur earlier in Zero Lag MACD. For trading the Zero Lag MACD, the same rules which apply to normal MACD also apply to zero lag MACD.

Regards

-Anant

oldmirage · #**924** 1st May 2007, 02:28 AM

This thread is running for long time and have missed lot of action in it but I had one basic question.
If I want to test the code posted by knowledgable ppl here, what softwares do I need to install ?
Are all or part of the software requires subscription ?

kkseal · #**925** 1st May 2007, 04:00 PM

Hi Anant,

Nice to see you back and great stuff as usual.

Haven't read your previous posts on MA lag (I too was out of touch for some time). Will hunt for them.

Regards,
Kalyan.

anilbalaji11 · #**926** 1st May 2007, 04:16 PM

Does technical analysis really work now a days or it is manipulated by big operators , fiis and mfs etc. I doubt , time has changed. Now a days atleast in intraday technical analysis sucess rate has come down due to the manipulation of prices by big operators etc

asnavale · #**927** 1st May 2007, 07:14 PM

Quote:

Originally Posted by kkseal

Hi Anant,

Nice to see you back and great stuff as usual.

Haven't read your previous posts on MA lag (I too was out of touch for some time). Will hunt for them.

Regards,
Kalyan.

Hi Kalyan,

Thanks for your encouraging words. My earlier post on MA lag is in this thread on page no. 86, post no. 851. Hope you will find it useful.

Regards

-Anant

asnavale · #**928** 1st May 2007, 07:21 PM

Quote:

Originally Posted by oldmirage

This thread is running for long time and have missed lot of action in it but I had one basic question.
If I want to test the code posted by knowledgable ppl here, what softwares do I need to install ?
Are all or part of the software requires subscription ?

Hi Oldmirage,

The codes posted in this thread are for AmiBroker. AmiBroker is not FREE software. However you can download a trial version from http://www.amibroker.com. The limitation in the trial version is that you can not save anything.

If you are using other software like metastock, fibotrader or fcharts etc. you have to translate these codes yourself or by somebody who can do it.

Regards

-Anant

oldmirage · #**929** 1st May 2007, 08:06 PM

thanks Anant.

I am throwing out another question for ppl who had participated in testing as well as writing different systems.

After improvising these systems, did anyone developed any quantamized parameter to measure the success rate of these system ?

I work with mathematical models and this kind of thing come to mind more naturally to me. If I can help this forum with my knowledge, please guide me what and where should I look.

Quote:

Originally Posted by asnavale

Hi Oldmirage,

The codes posted in this thread are for AmiBroker. AmiBroker is not FREE software. However you can download a trial version from http://www.amibroker.com. The limitation in the trial version is that you can not save anything.

If you are using other software like metastock, fibotrader or fcharts etc. you have to translate these codes yourself or by somebody who can do it.

Regards

-Anant

CreditViolet · #**930** 2nd May 2007, 11:14 PM

Talking of removing lag, how about zero lag

Just a side note about removing lag, the key thing is the Time-Series you are 'de-lagging', i.e the concept of your indicator should be a 'fundamental' truth about the markets before smoothing it. I hate to be the one to break it but you guys are going to get nowhere removing lag from a MACD or Stoch on a data series that is not even stationary in the first place.

Attached is an eg of how I remove 'lag' from a 'Sentiment' Indicator that I calculate real-time.

Similar Threads for: Experiments in Technical Analysis
Thread	Thread Starter	Forum	Replies	Last Post
Technical Analysis Course	snowy	Technical Analysis	28	7th November 2008 07:37 PM
Technical analysis course	sh50	Seminars & Tutors	18	10th June 2007 03:46 PM
Hello everybody.Technical Analysis	vijay chauhan	Introductions	2	30th May 2007 03:09 PM
Technical Analysis	crosscurrency	Forex	0	30th October 2005 11:00 PM