I'm not putting emphasis on the leverage but on the percentage. Indeed 1% is 1% whatever the leverage. I'm just saying that for this method 1% is just too much. Regarding the leverage: 1% on a 5 pip SL that's already 20:1 leverage (2 minilots for ...

Extracted Thread (per thread starter's request)

This is exactly it. You trade fixed size. More precisely piecewise constant lot size since you increase from time to time. While a lot of people including myself trade %. That's the reason I wanted to warn "1% is far too high with this method". You ...

Extracted Thread (per thread starter's request)

I never post in a thread I didn't fully read and understand. I think YOU totally missed my post. So please do yourself read and understand what I wrote. You didn't get it at all. Second you clearly didn't read the thread yourelf. Fez explained that ...

Extracted Thread (per thread starter's request)

Hi, I've been eagerly reading all the posts since this morning. When I was reading the millipede thread my only regret was that the OP was gone since long. I tried the millipede. Three times. I failed. Three times. The mistake I made: over leverage. ...

Extracted Thread (per thread starter's request)

Metatrader version? Metatrader? Oh no way! I wrote my own plateform url I also use R for quick stats. Broker used? FXCM and Dukascopy. For no other reason that they provide a java API from which I can download real time and historical tick data. ...

Poll: Which data are you using for developing your EAs?

Maybe there is something to be done with the distributions. The picture below shows an example of white noise and the histogram of the absolute value of the innovations on the left. On the right, a sine wave and the same histogram. The modes are on ...

Choppy market index: any good ideas?

I would like to clarify that the 50%-50% is not the reason for the convergence in distribution to a Gaussian (Central Limit Theorem is). Even with a 70%-30% biased coin it would converge; except that the mean would not be zero. The time bars (like ...

Choppy market index: any good ideas?

Can you develop this idea a little bit please? I like very much the idea of using probabilities but I can't figure out how you see this. BTW the distribution for 4 range bars is a little bit complex with 3 modes. But after 7 bars it becomes Gaussian ...

Choppy market index: any good ideas?

If you absolutely want to believe in this statistical arbitrage you should at least use the correct weights: 1.6 lots CAD/JPY for each lot EUR/JPY. Which pair to buy or to sell depends on if the regression is above or below zero. The cointegration ...

Correlated pairs method for disciplined traders

You made my day

Correlated pairs method for disciplined traders

ROFL. Both CAD/JPY and EUR/JPY are expressed in JPY: buying one and selling the other exactely cancels out the yen. This "system" is simply about randomly shorting EUR/CAD with no stop paying twice the spread. BTW that's now 12 weeks E/Cad can't ...

Correlated pairs method for disciplined traders

You have to cut inside the 1 min candles. The same 1 min candle can overlap two CRB because the CRB can exceed the range while the price is travelling from the 1min low to the 1min high. You have to interpolate the price movement inside the bar. You ...

Statistical mean of the market [quant corner]

Here is the next bar. Notice the range is between the low and the high(=close). The open price is not taken into account. image

Statistical mean of the market [quant corner]

No in the first IF I use the difference between the current price (price=tick.getPrice()) and the lowest received thus far in the forming candle (candle.getLow()). In the second I use the difference between the current price and the highest received ...

Statistical mean of the market [quant corner]

Because they interpolate from 1 min bars. They cut the bar anywhere without concern if there is a tradable tick here or not. A 5 pip gap can easily happen inside the 1 minute. I don't use brute force. I write the equations on paper then I code them. ...

Statistical mean of the market [quant corner]

I'm not especially concerned with execution(1). It is one possible use of this. I have other ideas. Originally I wanted to avoid adding to a position in such a market because the probability is high that the price retraces right after I have a ...

Choppy market index: any good ideas?

It is impossible per construction. The next bar happens when a tick appears outside of the range. If the tick escapes the range by the high, the close of the bar is at the high. A bearish bar closes on its low when a tick escapes the range by the ...

Statistical mean of the market [quant corner]

My favorite one it that you can acquire financial independence starting with a broken piggy bank because we all of us wish it were true.

What are your favorite trading myths?

Hi! I don't understand what you call fake or synthetic ticks. The ticks are the ones in your database they are real, aren't they? Something is wrong with your charts. Look at the first screenshot 32nd bar: it isn't possible that a green bar (up) has ...

Statistical mean of the market [quant corner]

hehe I know. The tick by tick distribution of the price (market profile) becomes very close to a gaussian. That's why I'm pretty sure that in this case it is possible to set a limit order at say the 2nd SD with a SL at the 4th in the direction a ...

Choppy market index: any good ideas?