Mathematical trading

I took a short,i shared that trade and people are free to take the trade in their accounts that means duplicate.I don't mind.But from sharing the trades to sharing the methods or things about the method is a long way.You don't see the difference between these 2 simple things?You said you are one hell of a mathematician,you cant see the logic difference?I have enough money,i don't need some stupid fees for my trades like you may belive.You can see my trading balance both in FX and in shares.i think I will survive without selling the trades.

You still avoid any pertinent questions like the plague. Which markets do you invest in? It is worth noting what was already said. Making some trades and posting them here does not make you a mathematician. You have not shared a single thing about your method. Stop complaining about it. Neither I nor @IFeelFree are asking about the ins and outs of your method. We want to know what field you are using specifically. To quote you, how can you not see the logical difference between those two things?
 
In what world you can trade EurUSD on NYSE?You seems a bit confused here.Why did I paid?For accurate datas from 2000 to 2014. 600 Gb worth of datas tick by tick.

I invest in FX(I tought is was clear since I said EURUSD) and you are supposed to be a trader.You should know what EU means and in which market it belongs.And US shares,you have my open trades on this very topic,search and see.

I never said that the EUR/USD was traded on the NYSE. You just implied that. Keep those blatant deflections coming. :LOL:

Nothing is ever clear with you. You behave like a cult leader. I did not ask why you paid for it. I asked why you overpaid for it. Do not misquote me.

US shares sound as though you are trading CFDs. That is elucidating.
 
I never said that the EUR/USD was traded on the NYSE. You just implied that. Keep those blatant deflections coming. :LOL:

Nothing is ever clear with you. You behave like a cult leader. I did not ask why you paid for it. I asked why you overpaid for it. Do not misquote me.

US shares sound as though you are trading CFDs. That is elucidating.

you have no clue about what historical datas are,what are those good for,what markets exists and pretty much you know nothing. But you just wanna know my method,that is your goal.Wasnt clear where I invest from my first statement?You have no clue about markets,this is the problem.Do yourself a favor,stop asking stupid questions.

" Making some trades and posting them here does not make you a mathematician. You have not shared a single thing about your method. Stop complaining about it. "

You and FeelFree are the ones crying here because I don't talk about my method.I am not complaining.Stop whining about that.Posting some trades is the reason why forums exists.Not for talking about all sort of useless methods and theories that bring no money.I am a practical person,you have examples of that.You just talk.Again,i am sure you have no knowledges about markets and trading,you just like to talk.
 
You know what that stands for right? :whistling

How amusing that you chose something that userque already listed and used his abbreviation.

Yes,amusing that I use Support vector machine. I am glad you find something funny.Simple minds always find various things amusing.
 
I paid for a custom one.
That's what I suspected. It's some kind of commercial machine learning/neural net program you purchased and have adapted to trading. You probably trained it on your dataset and are using it to predict market prices.

The problem is that you don't really know what it's doing, and you have to try to figure out empirically under what circumstances it might forecast market prices. It sounds kinda shaky to me. It might kinda work for a while and then stop working as market conditions change. The fact that it only allows infrequent trades suggests that its not very robust.
 
Last edited:
That's what I suspected. It's some kind of commercial machine learning/neural net program you purchased and have adapted to trading. You probably trained it on your dataset and are using it to predict market prices.

The problem is that you don't really know what it's doing, and you have to try to figure out empirically under what circumstances it might forecast market prices. It sounds kinda shaky to me. It might kinda work for a while and the stop working as market conditions change. The fact that it only allows infrequent trades suggests that its not very robust.

Yes,probably.
You don't have to trade what I post. Chen will give you 50% year by September.We will see by then what you have and what I have.
And also the problem is that you don't really know what I do and that is the big problem. I know perfectly what I do.Again,i am sure you aren't profitable.You just talk.Talking is fine,some people have only this left.

the other ''sharp'' fellow you work with is Hhiusa? I don't think you gonna crack that method too soon,if this is the case.
 
Yes,probably.
You don't have to trade what I post. Chen will give you 50% year by September.We will see by then what you have and what I have.
And also the problem is that you don't really know what I do and that is the big problem. I know perfectly what I do.Again,i am sure you aren't profitable.You just talk.Talking is fine,some people have only this left.
After 8 months, I have a 32% profit on a $75,000 starting balance, which when annualized is a bit over 50% yearly. We'll see what it is in September. I wanted to give it a year of living trading before I made a fuss about it. Keep in mind, I only trade once a week. That's about an hour a week, and then have the rest of my time free. I don't want to spend all my time staring at charts.

the other ''sharp'' fellow you work with is Hhiusa? I don't think you gonna crack that method too soon,if this is the case.

No, it's not Hhiusa. The guy I'm talking with has already developed his program and he's just trying to optimize it. He uses a similar approach as me but a different implementation and a different portfolio. I won't share my portfolio with him because I don't want it to get out to others (some of whom might have deep pockets) what I'm trading so that they could duplicate my trades. As soon as any trading strategy becomes too popular, it stops working because others front-run it.
 
Entertaining thread :) . I don't see what all the fuss is about. OP is happy he's found a system that works for him and wishes to post his trades here to show off his results. The stubbornness to reveal anything remotely specific about the methodology is understandable, I'm not sure why you keep trying to force it out of him.

@t.roland here's some advice: I'm not certain what you're trying to achieve but if you're looking to gain respect in a community for the work you do, you might want to adopt a slightly different attitude. This is a friendly forum filled with really helpful and friendly people -- the reason people "attack" you is because you're neither friendly, helpful nor interesting. You don't add value, basically. I'm sure there are plenty of lesser communities out there that would happily absorb your current attitude and personality but I doubt you'll find much respect around here.

Just my 2¢..
 
Entertaining thread :) . I don't see what all the fuss is about. OP is happy he's found a system that works for him and wishes to post his trades here to show off his results. The stubbornness to reveal anything remotely specific about the methodology is understandable, I'm not sure why you keep trying to force it out of him.

Totally agreed. Whatever works for you best, right? Besides, it's everyone's own business what and how he does for profitable trading...
 
Entertaining thread :) . I don't see what all the fuss is about. OP is happy he's found a system that works for him and wishes to post his trades here to show off his results. The stubbornness to reveal anything remotely specific about the methodology is understandable, I'm not sure why you keep trying to force it out of him.
My reaction to @t.roland's claim was skepticism when he said that he "created a purely mathematical trading method with a extremely low chance of failure", and yet he wouldn't (or couldn't) answer the most basic questions. I don't mind people being secretive about their system, but I sensed that he wasn't being honest with us. A true algorithmic trader can talk about his system without giving away any secrets.

After repeated questioning, he finally revealed that he was using commercial "machine learning" (probably neural net) software, into which he input market data and had it forecast optimal trades. That's fine. If it works for him, great. But he gave the impression at the beginning that he was a mathematician and wouldn't deny it when I asked him. My sense is that he doesn't understand what his system is doing. That can be dangerous. It might work under certain market conditions, and when market conditions change, it can fail. Every neural net system has a finite lifetime, after which it stops working. Also, my impression from reading the literature is that trading profits from neural net approaches are modest, at best. For example, see this article on Neural Networks from Investopedia, in which they claim that "a 10% increase in efficiency is probably the most you will ever get from a neural network."

I wish @t.roland the best with his system, but I don't think that he understands the "complex mathematics" he's using. Many traders have tried and failed trading with neural net algorithms when they treat them as "black boxes."
 
Agree with all three of you, I would never trust my money with anything I hadn't built myself, from scratch. I have my own suspicions about the guy's actual proficiency with these sorts of things but I'll keep them to myself ;) .
 
My reaction to @t.roland's claim was skepticism when he said that he "created a purely mathematical trading method with a extremely low chance of failure"

All curve fits are perfect. There's no need to be sceptical. It's something that all beginners get into, and all trainers show their results with.

The question then becomes: does the real life entity that sell you the assets to speculate respect those curves ? They don't. They always aim to buy back cheaper than what they sold at. In effect, in practice all curve fits are defeated.

To test the fragility of any curve fit, you merely need to place a tiny real trade of insignificant value. It will destroy the curve in no time. I noticed some curve fitters prefer a longer term play, but they are merely delaying the results. The outcome will be the same.
 
Last edited:
Top