Trading algorithmically a folio without stops (with IB), real $$$

[AdrianHagh81]

Lol,

On the spectral dilation thing, I have to agree with full auto trading.

The spectral analysis posted by jcl doesn't really mean anything.

However having said that, i do notice that the prices or more specifically the price changes/returns
on some asset class do exhibit repeatable patterns on very short time frame that can be marginally exploited.
This may be due to the specific market microstructure.


But this is a fun thread.

FYI, my credentials in case anyone ask
Phd EE. Altough my h-index sucks lol.


new to trading, just recently been profitable using automated futures strategy
based on volatility.

 

[fullautotrading]

Lol,

On the spectral dilation thing, I have to agree with full auto trading.

The spectral analysis posted by jcl doesn't really mean anything.

However having said that, i do notice that the prices or more specifically the price changes/returns
on some asset class do exhibit repeatable patterns on very short time frame that can be marginally exploited.
This may be due to the specific market microstructure.

But this is a fun thread.

Yes,

Glad you are having fun AdrianHagh81.

(And btw there is no need of credentials here: common sense and intellectual honesty are largely sufficient

I agree that there are situations that can be exploited (I would not talk of "repeatable patterns" though). That is in fact the rationale of introducing the mechanism, among several other ones, I called it "strong SDX override" (described earlier in a post) to capture temporary price accelerations. This can be done optimally through automation, and I currently do that. Manually, it can be harder and quite stressful (also due to the fact that they are not very frequent events), however an algorithmic platform can follow the rapid move very accurately through a fast paced pyramiding and close most of the profitable players before a reversal (those which remain "in loss" can be taken care later, as we keep that information in memory). These mechanisms are however just like the "icing on the cake", a refinement, they can't really be the "pillars" of a trading methodology. However, sometimes (but not much frequently, must say) do succeed in "stealing" away nice chunks of profit (lately saw that on TZA, for instance).

 

[AdrianHagh81]

Yes,

Glad you are having fun AdrianHagh81.

(And btw there is no need of credentials here: common sense and intellectual honesty are largely sufficient

I agree that there are situations that can be exploited (I would not talk of "repeatable patterns" though). That is in fact the rationale of introducing the mechanism, among several other ones, I called it "strong SDX override" (described earlier in a post) to capture temporary price accelerations. This can be done optimally through automation, and I currently do that. Manually, it can be harder and quite stressful (also due to the fact that they are not very frequent events), however an algorithmic platform can follow the rapid move very accurately through a fast paced pyramiding and close most of the profitable players before a reversal (those which remain "in loss" can be taken care later, as we keep that information in memory). These mechanisms are however just like the "icing on the cake", a refinement, they can't really be the "pillars" of a trading methodology. However, sometimes (but not much frequently, must say) do succeed in "stealing" away nice chunks of profit (lately saw that on TZA, for instance).

Hi Professor Gastaldi,

I've just read your paper on fuzzy linear regression.

Very intresting stuff. Are you using some fuzzy linear for Gbot ?

 

[AdrianHagh81]

"It is principially impossible for any trading methodology to statistically cope with a random walk price curve. The existence of such a methodology would cause logical paradoxes, such as generating money out of nothing."

Real price curves are doubtless generated by stochastic processes, but they are no random curves, otherwise no trade system would ever work. F

In a 'real' random curve the spectrum would be flat.

Interesting,

Well if you apply an Ito integral to a brownian stochastic process from
my understanding the only non random variable would be the upper and lower
limit of it.

 

[fullautotrading]

Hi Professor ...

I've just read your paper on fuzzy linear regression.

Very intresting stuff. Are you using some fuzzy linear for ... ?

No professor here please, I am just one who is making a journey, kindly supported by enlightened investors, and trying to humbly build up something new and that both makes sense rationally and works, with all those of you who are interested

Thank you for the thought, anyway: I am using no "fuzzy" concepts in this context...

... when $$$ is concerned, we need to be very "crisp" ;-)

 

[AdrianHagh81]

No professor here please, I am just one who is making a journey, kindly supported by enlightened investors, and trying to humbly build up something new and that both makes sense rationally and works, with all those of you who are interested

Thank you for the thought, anyway: I am using no "fuzzy" concepts in this context...

... when $$$ is concerned, we need to be very "crisp" ;-)

Right Sir,

Well I just thought maybe before people tried to give you lessons on probability distribution they need to know what field you are an academic of. Lol

Good luck on your journey.
And in case you are short of grad students for number crunching tedious stuff,
I'll be happy to help.

 

[jcl366]

The spectral analysis posted by jcl doesn't really mean anything.

Would you be so kind and explain that?

Spectral analysis is used to determine dominant cycles in price curves. When you look at the spectrum, you can even identify the weekly cycle, at 120 hours. Although cycles are normally nonstationary, they exist long enough in some assets, especially indices, to be exploitable. We're doing that, among other trade methods. When you get hard cash with it, why does it not mean anything?

 

[jcl366]

I can't give private lesson here in Statistics. Especially, to hostile people not minimally interested in learning, but only to systematically twist and purport

I am not hostile, in fact I even have some sympathy for your approach. Admittedly you don't make that easy with your permanent personal attacks on people with a different opinion.

I am not saying that your system is principially bad. New grid trading methods became very popular recently. We have even such a system on our website - just look for "Zorro Z4". It's free. Some guys from Denmark also run a system similar to yours - look for "Hourglass system". But there is a difference: all those systems make good profit. Ours returned 170% since its release last year. The users more than doubled their capital so far. Therefore, you seem to be doing something wrong. Maybe it's the hedging mechanism, or it's your assumption of a "statistical dominance" where none exists. Unfortunately you don't listen to advices.

The market gives us only a small edge. Any little mistake turns a winning system into a losing one.

 

[AdrianHagh81]

Would you be so kind and explain that?

Spectral analysis is used to determine dominant cycles in price curves. When you look at the spectrum, you can even identify the weekly cycle, at 120 hours. Although cycles are normally nonstationary, they exist long enough in some assets, especially indices, to be exploitable. We're doing that, among other trade methods. When you get hard cash with it, why does it not mean anything?

Right, probably have something to do with options expiry on friday ?

But according to myfxbook your spectral filtering method z12 is loosing money this year.

And your grid method is showing 1600 pips profit but -900 pip loss in open positions.

What real hard cash are you talking about ?

 

[fullautotrading]

I am not hostile, in fact I even have some sympathy for your approach. Admittedly you don't make that easy with your permanent personal attacks on people with a different opinion.

I am not saying that your system is principially bad. New grid trading methods became very popular recently. We have even such a system on our website - just look for "Zorro Z4". It's free. Some guys from Denmark also run a system similar to yours - look for "Hourglass system". But there is a difference: all those systems make good profit. Ours returned 170% since its release last year.

...

Your system returns 170% per year Ah ok, sorry I did not notice that. Well, then we have solved all our problems, I just need to download Zorro and we will immediately start recovering and becoming rich. That's brilliant.

I had already visited your site, don't worry, I am not so dumb.
What age are you btw ? If you are under 10-16, I promise to be very nice with you and don't treat you as an foolish scammer and troll.

Zorro: the system which steals from the rich and gives to the poor. ("The idea behind this software is simple: Let some wily machine redistribute money from the rich to the poor.")

Wow. That slogan must really attract a lot of wealthy investors. And it takes no much imagination to guess who the "poor" is