Trading Strategies Based On Digital Filters - page 69
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
pathetical wannabe
Simba,
I posted the link to M1 prediction on TRADE2WIN. I also posted links to results of some strategy which i'm using. So far you didn't post anything concrete here except one completelly repainted chart !!!
I have a feeling that you have nothing to post and you simple didin't realize
that SSA was cooking your results
Anyway, your philosophy of trading is different that mine, I was just curious
of your Goertzel results to compare with my Goertzel results to which i added some more funcions....if you don't want to post I can live with it
KrzysztofMy phylosophy of trading is different to yours because you are not a trader,you are just a pathetical wannabe that has bought all available commmercial systems , unable to create his own...
For being a clown that uses hyperfitting CSSA,you talk big words of small substance...
obsessed
As I said I can live without your results. I'm not going to discuss with you
anymore because you lose your temper too easy due to some reasons.
Regarding contest. I don't have time for bullshits like this, I don't have to prove anything to anybody.
I just concluded the facts that you don't post anything concrete here,
If you don't want that is your bussiness but dont try to manipulate the people to squeeze info from others because this is what you are doing.
Krzysztof
P.S. In which university they are teaching such language and giving MBA at the same time ???The one in Barcelona
...I presume you know which one is it,or at least which country is Barcelona in...
Which one did you go to?Krakowia night school for Sour Nerdies?
post 663 M1 prediction link is there. I'm very sorry to say but you are not a partner for conversation with me anymore.....unless you will post screenschoot of your MBA diploma here
Edited by Simba,Information not relevant for the thread
Just be carefull not to get repainted spectrum
Krzysztof,
you seem to be obsessed by the repaint issue. I guess you must have lost a lot of money with a misplaced interpretation of what a non-causal filter was telling to you.
Any time you analyze data, as you have a new tick arrival, the whole analysis changes. And it can be a big change. If you are doing spectral analysis, whatever the method you are using, be it mesa, fft, goertzel or whatever, every new tick changes the analysis itself. I mean, the whole world is repainting every millisecond, do you know greek philosopher Heraclitus? He says "panta rei", everything changes.
And that's what market is: change.
So the real problem for a trader is how to make money with change of price as time flows. And I believe that the first important thing is to put the right frame to the temporal horizon, as what is probable in the next 2 seconds is different from what is probable in the next 2 days. That's why SIMBA is right at pointing out your 'timeframe mess'.
Then,
if SSA or Hodrick-Prescott non-causal filters, which are not telling you to buy or sell, although your NN software may understand so , help in getting a sharper peak analysis with mesa or goertzel, like they do (see pics : on the left 'raw prices', on the right 'denoising with ssa'), I'm not scared of using them.
Panta rei
Krzysztof,
you seem to be obsessed by the repaint issue. I guess you must have lost a lot of money with a misplaced interpretation of what a non-causal filter was telling to you.
Any time you analyze data, as you have a new tick arrival, the whole analysis changes. And it can be a big change. If you are doing spectral analysis, whatever the method you are using, be it mesa, fft, goertzel or whatever, every new tick changes the analysis itself. I mean, the whole world is repainting every millisecond, do you know greek philosopher Heraclitus? He says "panta rei", everything changes.
And that's what market is: change.
So the real problem for a trader is how to make money with change of price as time flows. And I believe that the first important thing is to put the right frame to the temporal horizon, as what is probable in the next 2 seconds is different from what is probable in the next 2 days. That's why SIMBA is right at pointing out your 'timeframe mess'.
Then,
if SSA or Hodrick-Prescott non-causal filters, which are not telling you to buy or sell, although your NN software may understand so , help in getting a sharper peak analysis with mesa or goertzel, like they do (see pics : on the left 'raw prices', on the right 'denoising with ssa'), I'm not scared of using them.Rich,
Congratulations for your very detailed analysis
By looking at your graphs we can see there are 3 or 4 near constant cycles,each one of them seemingly having small variations around the individual "nominal period"/"nominal frequency" axis...Is that so?
Regards
Simba
Krzysztof,
you seem to be obsessed by the repaint issue. I guess you must have lost a lot of money with a misplaced interpretation of what a non-causal filter was telling to you.
Any time you analyze data, as you have a new tick arrival, the whole analysis changes. And it can be a big change. If you are doing spectral analysis, whatever the method you are using, be it mesa, fft, goertzel or whatever, every new tick changes the analysis itself. I mean, the whole world is repainting every millisecond, do you know greek philosopher Heraclitus? He says "panta rei", everything changes.
And that's what market is: change.
So the real problem for a trader is how to make money with change of price as time flows. And I believe that the first important thing is to put the right frame to the temporal horizon, as what is probable in the next 2 seconds is different from what is probable in the next 2 days. That's why SIMBA is right at pointing out your 'timeframe mess'.
Then,
if SSA or Hodrick-Prescott non-causal filters, which are not telling you to buy or sell, although your NN software may understand so , help in getting a sharper peak analysis with mesa or goertzel, like they do (see pics : on the left 'raw prices', on the right 'denoising with ssa'), I'm not scared of using them.Beautifull pictures. No I'm not obsessed with repaint, if you know what you doing that is fine you must be just carefull as I wrote. Also posting such pictures without explanation is very misleading.
Regarding mess. There is no mess at all here. I gave prediction using H1/H4 to align with Simba methods and also posted the link which proves that stabile cycles on M1 exist. Two independent things.
Krzysztof
repaint
I believe you should be more precised. Every new tick will change the analysis
related to FUTURE and CURRENT data but in case of SSA/HP will change HISTORY data in your system. How your system will act on this that's another story. Or I'm wrong ??
Krzysztof
Rich,
Congratulations for your very detailed analysis
By looking at your graphs we can see there are 3 or 4 near constant cycles,each one of them seemingly having small variations around the individual "nominal period"/"nominal frequency" axis...Is that so?
Regards
SimbaYou are right. The first row of peaks from the left is starting at 54 and slowly shifting towards 65 as you can also see in picture without the 3d elevation. The second peak is really very stable near 39-40 for all the observations. Keep in mind that I am using a relatively short window of 300 bars and that from the first observation (lower row of 2d pic) to the last (upper row) there is a 200 bar shift, that means that the peak remains stable for a time lapse of the same order of the window of obervation of past bars.
I also attach a text file with all peaks/valleys from the start to the end of simulation.
Thank you Simba.
Top of the morning to you, and I'm off with my new knife. (/swings it wildly in the air) Please reserve my gazillionaire club parking space, as I shall soon return with untold riches.
I shall change from
to 
.
/rides off on horse
I think you missed the point. Obviously past data cannot be changed by new data.
Whatever you call it as 'perspective' or as I called 'history data' result is the same in certain trading platforms - repaint like in post 633 including faking of trading strategy results due to future leak.
The advantages of using SSA and HP are well known but special versions are build to be used in real time trading.
Krzysztof