Trading Math - Part I

[trade4ever2day]

Quote from DontMissTheBus:

It's hard for him to do that: he doesn't actually understand the math he googled and posted, nor the subject that was being discussed when he interjected.

I'm done discussing this topic with that fella; He's clearly unhinged and I'm just wasting my time.

I think you are right. I also noticed he jumps from thread to thread insulting people. He never gives straight answers but his responses are always preceded by an ad hominen attack.

 

[trade4ever2day]

Quote from kut2k2:

... for those of you wondering whether [uncertainty of ruination] is true of all positive-expectation systems, the answer is yes. There is no closed-form proof for the general case, but there have been Monte-Carlo simulations. This can be found with a google search.

Yes, unicorns exist, they can be found with the help of some fantasy...

 

[kut2k2]

Quote from alexandermerwe:

Yes, unicorns exist, they can be found with the help of some fantasy...

See? You ask for a straight answer, then you act like a mule's ass when you get it.

You've proven to be nothing but a waste of time. Ignored.

 

[bwolinsky]

Quote from kut2k2:

Another idiot heard from. Where did I claim the proof was my own? If you or any of the other google-challenged slackwits reading this had any clue, you'd have found this proof yourself pages ago.

For the record, the theorem about the gambler's ruin was proven by Christiaan Huygens.

As I already stated, this is old stuff. That was a huge hint to all but the google-challenged slackwits that infest ET.

You did not cite anybody, thus, exactly what it is: plagiarism.

If you'd like to learn about citations, please type c-i-t-a-t-i-o-n into your google webbrowser.

 

[bwolinsky]

Quote from kut2k2:

Thanks for quoting bill's rant. He knows I have him on ignore so I don't have to directly read his idiocy anymore.

But this does allow the opportunity for a recap:
Notice the lack of any restrictions of any kind on the trading system, the underlying, or anything else. The statement is essentially that:

ALL trading systems under ALL circumstances will eventually fail unless stopped by a target profit or a time limit.

The challenge is to prove or to disprove the above statement.

So for those of us who actually passed our math courses, we remember that a SINGLE counter-example to a universal statement will DISPROVE the universal statement.
Remember: a single counter-example serves as a disproof.

I then posted the counter-example:
So now the imbecile is whining and nitpicking over the win-loss ratio above being 1. Seriously? We're supposed to take this bullshit seriously when a blanket challenge about a universal statement was issued? Now suddenly there are restrictions attached?

Like I said before, this isn't really about math, it's about a preening pedant presuming he's bringing some "new insight" to this forum.

However, for those of you wondering whether this is true of all positive-expectation systems, the answer is yes. There is no closed-form proof for the general case, but there have been Monte-Carlo simulations. This can be found with a google search.

There's no reason for you to be involved in this thread if you are going to ignore the OP. Seriously, why don't you quit pretending or acting like you know or understand this subject?

Nobody with 2 ounces of brains would splatter a thread with a case by case proof without citation or any original discussion. You clearly do not understand anything we are doing.

Also, I did not have to guess that all you did was google it and found some numbers and text to post without citation nor any explanation that would lead any average joe ET reader to come to the conclusion that you know anything about this subject.

Quote from alexandermerwe:

Obviously kut2k2's problem formulation didn't account for total profit stop or an exit time so it is an answer to (2). If q < p and win = loss then the probability of ruin is >0.

However, even that doesn't answer (2) completely. in (2) it is asked whether P of ruin < 1, not if P of ruin > 0. Obviously, 1 > 0.

As the starting capital grows large at every step of the game, P of ruin goes to 1.

In case (2) ruin is certain as T goes to infinity, something that has been answered already by DontMissTheBus I think. Thus it seems that kut2k2 did not provide an answer to the first question but stated the well-known fact that if a trading startegy has an edge p> q then it has positive expectancy if win = loss.

That is what you proved kut2k2. I don't think everyone else is an idiot and maybe you should stop insulting all ET members here because it is impossible everyone else is wrong and you are correct.

I thought both cases depend on the limit as T gets large, so if that's the answer to 2 I still don't see why it cannot also be the answer to 1.

 

[bwolinsky]

Quote from alexandermerwe:

Can't locate it or you may think you answered. So pls provide a link or explain.

He posted the gamblers ruin problem thinking that that was what we were talking about, so even though we probably do have a completed case-by-case proof of Gambler's Ruin, it does not in any uncertain terms answer any of the questions IntradayBill has asked.

@IntradayBill Do you have any reason for asking these questions? I guess I'm not sure what reaching an answer to these questions can do for you or with any practical application in finance. Why did you ask these 2 questions?

 

[Mike805]

Quote from intradaybill:

Statement 1:

For any trading system, with any parameters, if there is no target quit time or target profit after which it quits, eventually there will be total ruin.

I invite all the brains here to either prove or disprove Statement 1.

...and learn something...

Logically flawed question.

A trading system can enter and exit at will. This also means there exists a system that may *never* see conditions for entry. Statement disproved.

A trading system can have *any* parameters. This also means that there exists a system that has no parameters. Statement disproved.

For systems with parameters, entries/exits, and positive expectancy, chance of ruin is directly proportional to position size versus account size. Targets and time *CHANGE* the expectancy, hence, you've made a circular argument.

What a lame question.

Let me give anyone reading this thread some advice:

1. Trading is not about doing well in statistics class. It's about applying common sense to statistics. This is something which has yet to occur to Bill (or Bwolinsky) since their posts focus on trite details of a fundamentally flawed concept.

2. Common sense is elusive, especially to the stubborn ones.

3. Math is financially dangerous as it can contradict common sense.

 

[HATEtheRisk]

Quote from intradaybill:

Statement 1:

For any trading system, with any parameters, if there is no target quit time or target profit after which it quits, eventually there will be total ruin.

I invite all the brains here to either prove or disprove Statement 1.

...and learn something...

Yes, absolute correct.

For a market timer like me. But likely not for an Trend follower.
I dont know, i never was interestend in Trend Following.
I just want to get in and out of the trade.

Of course, if you have no target, you have no reason for the trade in the first place.

And the odds on your side are only there for a specific time.

Odds are changing all the time.

Market Timing Cycle Trading, how i do it, is all about odds, its like a chess game or Poker.

The mathematic here is A+B+C = D (what is enter)
-> if B changes, you must be on guard and manage the risk.
-> if A + B or B+C change against you = Exit all !!!

There is no really math in trading. All you need is %, that you can calculate your risk and expected profit.
----------------------
But the trend followers are having, like always a totaly different approach, with trading a win rate of <40% and a risk/reward of 1 : 20 !!!

What ever, i make more money and i make it quicker and without less stress.

PEACE

 

[bwolinsky]

Quote from HATEtheRisk:

Yes, absolute correct.

For a market timer like me. But likely not for an Trend follower.
I dont know, i never was interestend in Trend Following.
I just want to get in and out of the trade.

Of course, if you have no target, you have no reason for the trade in the first place.

And the odds on your side are only there for a specific time.

Odds are changing all the time.

Market Timing Cycle Trading, how i do it, is all about odds, its like a chess game or Poker.

The mathematic here is A+B+C = D (what is enter)
-> if B changes, you must be on guard and manage the risk.
-> if A + B or B+C change against you = Exit all !!!

There is no really math in trading. All you need is %, that you can calculate your risk and expected profit.
----------------------
But the trend followers are having, like always a totaly different approach, with trading a win rate of <40% and a risk/reward of 1 : 20 !!!

What ever, i make more money and i make it quicker and without less stress.

PEACE

Might have been the quickest profits on my sim trades before I go live with a 60 cent profit on GC in less than 15 seconds. And it is trend following price physics and it does get in and out all day long and there aren't any monthly losses so close drawdown is zero, win percentages trend following wise are 85-99+%, not the 40% you've seen, and I don't see any system with risk/rewards of 1:20, the profit factor is above 15 on the longs, but the return on Max Portfolio drawdown is a whopping 275 times.

No, I doubt a trend following system can get much better. You should really look into it.

 

[virtualmoney]

Quote from oldtime:
this exactly why we don't let people with high IQ's run the world

Especially if some turn out to be the same person debating furiously against oneself.:eek: