Is there a need for a T.A.? - page 23

 
FreeLance: Can you independently determine the percentage of "excellent" and above?

The IQ of excellent students (stateside) is around 135-140 and above.

The "normalization condition", which defines the s.c.o. of the curve, says that at IQ=90 the integral distribution function is 0.25, and at IQ=110 it is 0.75. It is easy to estimate that for a Gaussian distribution, the value of the integral function is 0.75 at a point approximately equal to 0.6745 sigma. Hence, the sigma of the IQ bell is approximately 10/0.6745 = 14.8.

So, excellent students would be on the right side of 2.36-2.70 sigma. Well an average of 2.53 sigma. The i.f. value at this point is 0.9943. Hence the answer: excellent pupils and steeper pupils are about 0.57%.

That's not much, in fact there are more of them. The Americans overestimate the IQ of their excellent students. Or it is simpler: it is not necessary to have high IQ to be an excellent student. In fact, they are really only correlated concepts, but not identical.

 
Mathemat:

The IQ of the (state) excellent students is about 135-140 or higher.

The "normalization condition", which defines the s.c.o. of the curve, says that at IQ=90 the integral distribution function is 0.25, and at IQ=110 it is 0.75. It is easy to estimate that for a Gaussian distribution, the value of the integral function is 0.75 at a point approximately equal to 0.6745 sigma. Hence, the sigma of the IQ bell is approximately 10/0.6745 = 14.8.

So, excellent students would be to the right of 2.36-2.70 sigma. Well an average of 2.53 sigma. The i.f. value at this point is 0.9943. Hence the answer: excellent pupils and steeper pupils are about 0.57%.

That's not much, actually there are more of them. The Americans overestimate the IQ of their excellent students. Or it is simpler: it is not necessary to have high IQ to become an excellent student. In fact, they are really only correlated concepts, but not identical.


It was basically about evidence and this approach is speculation, the brain in general is well adapted (optimised) for speculation, for simple and quick conclusions, for everyday life
 
Mathemat:

The IQ of the excellent traders (stateside) is around 135-140 and above.

The "normalization condition", which defines the s.c.o. of the curve, says that at IQ=90 the integral distribution function is 0.25, and at IQ=110 it is 0.75. It is easy to estimate that for a Gaussian distribution, the value of the integral function is 0.75 at a point approximately equal to 0.6745 sigma. Hence, the sigma of the IQ bell is approximately 10/0.6745 = 14.8.

So the honors traders will be to the right of 2.36-2.70 sigma. Well an average of 2.53 sigma. The i.f. value at this point is 0.9943. Hence the answer: excellent pupils and steeper pupils are about 0.57%.

That's not much, in fact there are more of them. The Americans overestimate the IQ of their excellent students. Or it is simpler: it is not necessary to have high IQ to be an excellent student. In fact they are really only correlated concepts, but not identical.

So it is with success in speculation...

Overstated about 10% and even 5% and 3%.

A good speculator is like a good fisherman or hunter.

I wonder what percentage of the time a fisherman pulls his rod and fishes, and what percentage of the time he is wasting air?

Probably no more than 0.57% of the time on average...

;)

----------

As an alaverde, an example of testing a "a la Lovina" trading strategy, but with rare trading signals...

Strangely, there are so few trades on m15.

;)


Strategy Tester: Lovina
Strategy Tester Report
Lovina
FXCM-Demo (Build 226)

SymbolEURUSD (Euro vs US Dollar)
Period15 Minutes (M15) 2010.02.08 19:30 - 2010.08.27 20:59 (2009.10.05 - 2010.10.15)
ModelBy open prices (only for Expert Advisors with explicit bar opening control)
ParametersPOINT_TakeProfit=1995; K_SLoss=451; ProfDreams=-10000; Max_Orders=24; Luft=1; Tolerance=1; Kz=25; Mz=220; BBack=220; SlipPage=2; K=1.5; KTrends=1.7; H=1.5; Lots=0.3; Alfa=4; M=10; TPSL=false; T_Expir=false; HV_Lag=11; grznt=4; FastLimit=0.07; SlowLimit=0.005; win=44; weith=0.2; betta=0.25; BarOnly=true; xFile=false;

Bars in history13790Modelled ticks27480Simulation qualityn/a
Chart mismatch errors0




Initial deposit50000.00



Net profit349629.17Total profit642312.65Total loss-292683.48
Profitability2.19Expected payoff2093.59

Absolute drawdown26632.41Maximum drawdown131924.01 (30.98%)Relative drawdown61.61% (37496.61)

Total trades167Short positions (% win)82 (68.29%)Long positions (% win)85 (57.65%)

Profitable trades (% of all)105 (62.87%)Loss trades (% of all)62 (37.13%)
Largestprofitable trade50336.49losing transaction-20978.03
Averageprofitable deal6117.26losing deal-4720.70
Maximumcontinuous wins (profit)13 (84653.24)Continuous losses (loss)6 (-57723.24)
MaximumContinuous Profit (number of wins)116501.42 (8)Continuous loss (number of losses)-57723.24 (6)
Averagecontinuous winnings5continuous loss3

So at 27480 ticks there are only 167 trading decisions.

The total is 0.6%.

 
FreeLance: So there are only 167 trading decisions per 27480 ticks.

The total is 0.6%.

Well, it is not at all the 0.6% they were talking about. In principle (this idea has been voiced here several times - and by C-4, and me, and someone else) there is a hope to build a good system, if you combine several average ones with high... er, can't remember... "stroboscopicity" (high ratio of full time to time in the marketplace).
 
Mathemat:
Well, it's not at all the 0.6% that was talked about. In principle (this idea has been voiced here several times - and by C-4, and me, and someone else) there is hope for building a good system if you combine a few average ones with high... er, can't remember... "stroboscopicity" (high ratio of full time to time in the marketplace).

I think that according to Mr. Taleb - ta!

Time is a ruthless integrator...

;)

 
FreeLance:
На семинарах часто спрашивают: «Каков процент успешных трейдеров из всех людей, приходящих на рынок?». Честно скажу, - не обладаю подобной статистикой. Некоторые авторы утверждают, что не более 10 процентов. Вполне возможно.

В знаменитой книге Н. Талеб «Одураченные случайностью» приводит замечательный расчет количества «успешных» трейдеров в зависимости от срока торговли на рынке.

Суть расчета в следующем: если, с вероятностью в 50% трейдер в течение года может либо уменьшить счет, либо увеличить, то через пять лет из 10,000 человек, пришедших на рынок, будет 313 трейдеров, ежегодно показывающих положительный результат!

Т.е. 3 % в нашем рассматриваемом случае.

Can you independently determine the percentage of "honours" and above?

how does this prove the correlation between profits and IQ?
 
FreeLance:

Avals 29.08.2010 09:26
FreeLance:

Вы может забываете, что за реальными сделками ( в основном;) стоят вовсе не спекулянты. И поэтому эти потери-выиграши для них - часть себестоимости/выручки (условий их бизнеса). И в итоге именно они несут бремя разностей.

Что касается чисто спекулянтов - статистика по рознице есть. она совпадает с распределением IQ. :)

я только о спекулянтах писал.

I still believe that the real economy feeds the speculators, not the other way round.

;)


it does not contradict the fact that the speculator's profit is a loss or shortfall in profits for other traders.
 
Avals:

This does not contradict the fact that a speculator's profit is a loss or shortfall in profits of other traders.

but always - any transaction is a service profit... and a cost for the participant.

But in the warped "speculative retail forex space" for the founders of a DC, the "speculator's" loss is their net profit. Too much even net profit.

Spreads can be reduced and bonuses on deposits, or bonuses can be added for active "trading" ;)

That's really "hilarious".

This is as close to forex as bookmaking is to the subject of betting.

 
FAGOTT:


If you create a TS which earns 10% per month on a dollar deposit in the real market - you will be rejected by any commercial bank! Even if it's 5%, they will kiss you all over the place.

Doesn't lose at any depot.


I have the simplest fixed lot Expert Advisor, slightly more complicated than Avalanche and mathematically sound, earning 5-10% per month. Places a non-limit order with SL and TP=SL or TP=2*SL. Drawdown for fixed SL can reach 25%-30%. If you diversify, i.e. run several systems with different SL parameters, the drawdown can be reduced to 15%. I can show testing for any parameters, for example SL=20 pips and TP=40 pips, I earn about 1000 pips per year, this is a good parameter, and 800 pips for eurusd.

Strategy Tester Report
R3
Alpari-Demo (Build 226)
Symbol EURUSD (Euro vs US Dollar)
Period 1 Minute (M1) 2009.05.01 00:00 - 2010.08.27 22:59 (2009.05.01 - 2010.08.29)
Model All ticks (most accurate method based on all smallest available timeframes)
Parameters bullpos=true; bearpos=true; lot1=0.1; SystemNumber=1; H0=200; B0=0; T0=2; H1=0; B1=0; T1=0; H2=0; B2=0; T2=0; H3=0; B3=0; T3=0; H4=0; B4=0; T4=0; H5=0; B5=0; T5=0; H6=0; B6=0; T6=0; H7=0; B7=0; H7=0; B8=0; T8=0; B9=0; T9=0;
Bars in history 489898 Ticks simulated 16269457 Simulation quality 25.00%
Chart mismatch errors 0
Initial deposit 10000.00
Net profit 1281.72 Total profit 8875.49 Total loss -7593.77
Profitability 1.17 Expected payoff 1.71
Absolute drawdown 97.20 Maximum drawdown 255.91 (2.38%) Relative drawdown 2.38% (255.91)
Total trades 750 Short positions (% win) 367 (31.34%) Long positions (% win) 383 (30.03%)
Profitable trades (% of all) 230 (30.67%) Loss trades (% of all) 520 (69.33%)
Largest profitable trade 38.60 losing deal -21.70
Average profitable deal 38.59 losing trade -14.60
Maximum continuous wins (profit) 4 (154.40) Continuous losses (loss) 12 (-157.08)
Maximum Continuous Profit (number of wins) 154.40 (4) continuous loss (number of losses) -195.40 (11)
Average continuous winnings 1 continuous loss 3

 
FreeLance:

But always - any transaction is a service profit ... and a cost for the participant in the process.

But in the warped "speculative retail forex space" for the founders of a DC, the "speculator's" loss is their net profit. Too much even net profit.

Spreads can be reduced and bonuses on deposits, or bonuses can be added for active "trading" ;)

That's a real "laughing stock".

It has as much to do with forex as bookmaking has to do with the subject of betting.


What's so funny? You don't like middlemen or the fact that they charge money for it? The intermediary (the shop) sells you food, the bank charges interest on your rent, the broker gives you the opportunity to

Or maybe it's just brokers that you don't like.

And what does this have to do with the speculator/non-speculator ratio?

Reason: