Experiments in Technical Analysis

kkseal · #**931** 3rd May 2007, 12:03 AM

Quote:

Originally Posted by CreditViolet

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.

Nice to see you on this thread CV.

You mean the fundamental deficiency lies where - is it in the concept behind the indicators or the data series itself ?

Regards,
Kalyan.

karthikmarar · #**932** 3rd May 2007, 01:10 AM

Oh...the Big BOSS himself on this thread... real surprise.

Nice to see you on this thread CV.

Kalyan, I suppose the basic deficiency is in Indicator itself. The Moving average SMA or EMA are Linear filters (Actually a Infinite impulse Response Filter- IIR) which work well on stationary or slow moving signals. However the price are nonstationary. Ofcourse nonlinear Filters like KAMA, VIDYA and Ehler's Filters respond much better to nonstationary data.

Now that throws up many interesting questions. Can we use the difference of two nonlinear filtered signals in a way similar to MACD? We will definitely explore this later when time permits.

Also many Indicator assume the price data to have a Gaussian Probability Distribution which is hardly true. Fisher Transform does help to approximate the price to a gaussian PDF.

Now one can ask, Does the fisher transformed indicator perform much better than ..say a conventional indicator like MACD? Is it really worth going through the hi-fi calculation? On a postional trader perspective, it may be a simpler to stick to the conventional MACD as my experiments reveal. I would definitely share more information and ideas on this ... again ..when time permits..

warm regards

Karthik

karthikmarar · #**933** 3rd May 2007, 01:44 AM

Quote:

Originally Posted by asnavale

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.

........ This is the procedure used to calculate the 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

Ananth

It is great to see you back in action.

Some analysis of the zero lag MACD. It is no way an attempt to deride your effort, just a critical Analysis..

It is true that the zero lag MACD is sharper and sooner in indicating the reversals / Turning points compared to the conventional MACD. However the zero lag MACD produces more signals and more whipsaws. Also many a times it indicates bullishness during sideways markets.

The peaks and troughs are much accentuated and thus over emphasizing the Bullishness or bearishness. The Histogram is much more jagged. In contrast the Conventional MACD is smoother and easier to work with.

Are we sacrificing too much to get a faster entry?

I put the same trading signals, which I used for the conventional MACD in the discussion “Trading the MACD”. I am enclosing two charts as examples.

Would like to have your views…

Warm regards

Karthik

kkseal · #**934** 3rd May 2007, 01:50 AM

Quote:

Originally Posted by karthikmarar

Oh...the Big BOSS himself on this thread... real surprise.

Nice to see you on this thread CV.

Kalyan, I suppose the basic deficiency is in Indicator itself. The Moving average SMA or EMA are Linear filters (Actually a Infinite impulse Response Filter- IIR) which work well on stationary or slow moving signals. However the price are nonstationary.

Yes, i thought so on reading of the nonstationary data series mentioned in CVs post. But he also mentions the Stoch. So maybe it's got more to do with the PDF aspect of the price data itself.

My word, this is getting fascinating! With now CV also joining in!! (Talk of the usefulness of Traderji!).

YEH DIL MANGE MORE ....

Regards,
Kalyan.

oilman5 · #**935** 3rd May 2007, 08:22 PM

dear karthik,
i hereby request u..to answer some question.as allof us know u r a great trader,system developer..with faith in KISS.PL REPLY TO A RECENT THREAD..TOOL OF A TRADING PRO'
Q. R PUT FOR UR READY REFERENCE
dont know,whether i am right to write here....forbidden art of trading will be exposed here.many a pro r available here in this great forum...i request them to make comments.
1. why they r superior ?
2. what time frame they trade?
3. how far they r subjective in trading ?
4. what r the reason for their actual consistency ?
5. on what condition they dont trade ?
6. what is their contingency plan[enough is enough..now i am booking loss]
7. what self sabotage u find most difficult to overcome..yet how u have done it ?
8. do u think beginner trader[4yr amateur] can be like u oneday ?if yes. HOW!

THANKS WITH REGARDS
OILMAN5

CreditViolet · #**936** 3rd May 2007, 10:12 PM

Ok, so we are talking DSP here.Talking abt filters, MAs are low pass filters while derivatives of SMA etc like MACD are hi-pass filters.But before we get into all that, lets talk about the time-series itself.

Ok, so a stationary series is one whose statistical properties are constant along its length. Its quite well documented that financial time-series are heavy-tailed and not stationary.Volatility clustering is another important topic, i.e tendency of autocorrelation in volatility.Given all these different conditions, I think its absolutely treacherous to try and detect signals in all this noise.Theres just too much randomness to deal with here than most people think and most are 'getting' fooled by it.

What I saw on this thread was a propensity to design 'buy and sell' systems but absolutely no effort was made into understanding the data itself. For me atleast, the buy-sell signal part hardly is the important issue, the key thing is to beat randomness.I have attached a map of my development process, the 'Analysis & Mining' is the primary area of my development. Another thing it does is to help with 'intuitive' understanding of markets. Ofcourse people are not interested in such undertakings, everyone wants to quickly get to the 'buy-sell' stage. Again, as I have said before, its not bad at all or something I complain about, as Napoleon once said - " Never interrupt an enemy when he is making a mistake". Ofcourse I am talking generally and not about you guys. The effort you guys have put in here is commendable but IMHO this is not the way to go around about system design.

CreditViolet · #**937** 3rd May 2007, 10:17 PM

Quote:

Originally Posted by kkseal

Nice to see you on this thread CV.

You mean the fundamental deficiency lies where - is it in the concept behind the indicators or the data series itself ?

Regards,
Kalyan.

The problem is indeed the concept. A MACD tells nothing but the difference b/w two MAs.It has nothing to do with trend at all, sure indicators can be useful but then its more about 'reading' it than anything else. A good example is Chick Goslin who uses the 3-10-16 oscillator beautifully.

Rgds

oxusmorouz · #**938** 3rd May 2007, 10:46 PM

Quote:

Originally Posted by CreditViolet

I have attached a map of my development process, the 'Analysis & Mining' is the primary area of my development.

Hi CV,
If you don't mind, can you please elaborate on step II (Determine type of targeted behavior, data processing and mining, 1st level testing and end result).

Obliged,
Oxymoron

syamu · #**939** 3rd May 2007, 11:05 PM

hi every one ..i need some clarification...can we import live data from odin work station in which we r placing orders?? suppose if i need data from any column say last traded price can i get that data so that i can use that in excel to do some other calculations

uasish · #**940** 3rd May 2007, 11:06 PM

CV,

Cant read properly,plz give another link (with magnified version).

Asish

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