a trading strategy based on Elliott Wave Theory - page 199

New comment
 
Taleb describes the collapse of a portfolio of many uncorrelated instruments. Roughly speaking, there was also automatic arbitrage (an Indian programmer/mathematician wrote a program to calculate the direction and size of positions). All profits for about three years (600 million), the trader lost in a few weeks. When he asked the Indian in despair - what is the probability of this scenario (what has already happened), he answered after calculation - 11(!) sigmas :)

Wings, wings.... legs
 
<br / translate="no"> In the case of a multi-currency portfolio, the amount of drawdown depends on the number of n instruments used as:
Dm(t)=SQRT(1/n)*t^(1-p).



Neutron, this would be true if currency pairs were independent. But I suggest to use their correlations and lot selection. In this case the drawdown may be significantly reduced. I think if we select currencies and lots so that the portfolio will move by 40-50 points, it will be a solution of the problem "how to outplay Forex" :)
 
About the multiple indicators. I tried that too. Not exactly that though. I considered as an entry signal the crossing of EMA(20) and EMA(40) on a sequence of timeframes from M1 to H4. I ran it all in the tester on EURUSD at M15, adding and removing timeframes with EMA. The result is interesting. M1 and H4 make the result worse. M5, M15, M30, H1 are improving. M15 and H1 especially sharply improve the result. As for indices on one timeframe... I think I will agree with most of them. They are all alike, and the main thing is that they all are derivatives of the price... Therefore they will hardly give us anything except confusion and calculation errors... In this respect I am surprised with another thing. Why are there almost no indices that take volumes into account? I do not agree that the volume in Forex is pure fiction. Just look at the volumes in MT4 and see how they periodically change during the day (here it is, by the way, a steady cycle!) And how their sharp growth correlates with a sharp jump in prices (not only on the news, by the way). When I was just starting out, I always used the growth of volumes during the jumps in the direction of the trend as an indicator of its reliability. Maybe the idea of multiple indicators can be somehow attached to volumes?
 

solandr волатильность так же как и фрактальность каждый понимает по своему :)))
Скажу по другому "амплитуда колебаний" у GBP/USD чуть больше чем у EUR/USD.
Чем чаще и "резче" происходят движения, как в одну так и в другую сторону, тем проще
увеличить или спустить депозит, нежели делать это при боковом тренде :)

I think that's what I'm talking about... The spread of the daily bars contains information about the maximum amplitude of price movements during the day. That is, we have information about the maximum potential profit of one trade during the day, which according to calculations should be equal for both pairs. Although, if you have passed in your strategy from the long-term analysis for 3-5 days ahead simply to the intraday game, taking into account the additional twitching of the pound within a day and to the orders with expected profits in 10-20 pips, then of course, the change of currency on the pound can only contribute to it. On the contrary, I try to avoid it, as it is probably easier to predict longer term trends (in the sense that you have more time to think and justify conclusions), rather than short term jerks in the currency during the day. I personally have nothing against your system though. Try to play on the pound. I wish you success with the new currency.

PS: By the way I can recommend you not to limit yourself to one of the currencies. I used to do it the same way - for example I looked at one EURUSD. But now I have installed my Expert Advisor on all available currencies at my broker. There are 19 of them. Correspondingly, I decreased the lot size, because the number of open positions has increased significantly. I think it helps me to observe some technical differences in movements of different currencies and improve the Expert Advisor accordingly. I also train my eye to recognize technical images and form my own suppositions, observations and conclusions which I think help me to improve my Expert Advisor according to the strategy described here.







solandr thanks for the tip :)))
I tried not to limit myself to one currency pair for a month.
Do you manage to track 19 currency pairs manually?
Or do you use your EA simultaneously on all pairs?
Share the secret:)))
 
Also, I doubt very much that any indicator will have the same p on all pairs. The nature of price movements is different, and I suppose p will be different too. In this situation it's better to choose the pair where p is maximal rather than diversify.



Yurixx, there should be only one indicator, which should work equally on all currency pairs and on all timeframes. Why create an indicator that works only on EUR/USD, and only on H1 timeframe?
If you have a choice - to trade on several Bp's at the same time 0.1 lot, or on one Bp of 1 lot,
I choose the first!
Take my word for it :)
 
<br / translate="no">
Also, I doubt very much that any indicator will have the same p on all pairs. The nature of price movements is different, and I suppose p will be different too. In this situation it is better to choose the pair where p is maximal rather than diversify.


Yurixx, there should be only one indicator, and it should work equally on all currency pairs and on all timeframes. Why create an indicator that works only on EUR/USD, and only on H1 timeframe?
If you have a choice - to trade on several Bp's at the same time for 0.1 lot, or on one Bp's for 1 lot,
I choose the first!
Take my word for it :)

I do. :-)
And about the indicator you just do not understand me. Of course there should be one indicator, I mean one and the same. The idea to create for each pair or for each TF an indicator of its own - only a person far away from Forex could come up with such an idea. I was talking about the predictive validity of the indicator, that is denoted by the letter p.
Take any, the best indicator and calculate its p on different pairs. You will see that these values differ from each other. If the indicator is really good, the difference will be small, 2-3 points. In the worst case it will be radical. But identical values on all pairs is science fiction.

I have checked both of them, and I think Rosh's position is more reasonable.
Yurixx no offence :)))

What's the grievance, Alex? Your point is confirmed by your own experiment. Quite respectable.
Also, I, on the same page, agreed with Neutron's arguments regarding diversification. So our positions are not so far apart.
 
Yuri, predictive validity itself isn't exactly clear what it is either. For example, muves behave well on a trend and lie on a flat. How are you going to calculate this very p? On average over the year? In the best or worst cases? By the way, the formulation of the question about the joint probability for a group of indicators does not seem quite correct to me.
 
Your point is confirmed by your own experiment. Quite respectable.


Yurixx new experiment with diversification and stop loss :)))


Ticket Open Time Type Lots Item Price S/L T/P Close Time Price Commission Taxes Swap Profit
9067137 2006.12.08 13:50 balance Deposit 5 000.00
9067143 2006.12.08 13:50 sell 2.50 audusd 0.7890 0.0000 0.0000 2006.12.15 10:53 0.7799 0.00 0.00 -106.00 4 550.00
9301680 2006.12.15 10:53 sell 5.00 gbpusd 1.9562 0.0000 0.0000 2006.12.15 17:00 1.9526 0.00 0.00 0.00 1 260.00
9301700 2006.12.15 10:53 buy 5.00 gbpusd 1.9566 0.0000 0.0000 2006.12.15 11:05 1.9574 0.00 0.00 0.00 280.00
9328406 2006.12.15 17:05 buy 5.00 gbpusd 1.9528 1.9530 0.0000 2006.12.15 17:49 1.9530 0.00 0.00 70.00
9328414 2006.12.15 17:05 buy 2.50 gbpusd 1.9528 1.9530 0.0000 2006.12.15 17:49 1.9530 0.00 0.00 35.00
9330359 2006.12.15 17:50 buy 5.00 gbpusd 1.9533 1.9620 0.0000 2006.12.19 10:08 1.9635 0.00 0.00 -25.90 3 570.00
9330369 2006.12.15 17:50 buy 2.50 gbpusd 1.9533 1.9621 0.0000 2006.12.19 10:08 1.9635 0.00 0.00 -12.96 1 785.00
9393202 2006.12.19 14:42 buy 11.00 gbpusd 1.9599 0.0000 0.0000 2006.12.19 14:47 1.9605 0.00 0.00 0.00 462.00
9393416 2006.12.19 14:47 buy 12.00 gbpusd 1.9608 1.9612 0.0000 2006.12.19 15:12 1.9623 0.00 0.00 0.00 1 260.00
9394181 2006.12.19 15:12 buy 13.00 gbpusd 1.9627 0.0000 0.0000 2006.12.19 17:01 1.9628 0.00 0.00 0.00 91.00
9396724 2006.12.19 17:01 sell 13.00 gbpusd 1.9625 1.9592 0.0000 2006.12.21 15:45 1.9579 0.00 0.00 -76.44 4 186.00
9455535 2006.12.21 15:46 sell 16.00 gbpusd 1.9575 0.0000 0.0000 2006.12.22 19:25 1.9566 0.00 0.00 -23.52 1 008.00
9487689 2006.12.22 22 22:57 sell 17.00 gbpusd 1.9586 0.0000 0.0000 2006.12.26 14:44 1.9581 0.00 0.00 -49.98 595.00
9491588 2006.12.26 14:45 buy 17.00 gbpusd 1.9578 1.9630 0.0000 2006.12.28 15:55 1.9644 0.00 0.00 -179.69 7 854.00
9535446 2006.12.28 15:56 buy 22.00 gbpusd 1.9651 0.0000 0.0000 2007.01.02 17:43 1.9731 0.00 0.00 -175.56 12 320.00
9608161 2007.01.03 21:25 buy 5.00 usdzar 7.0150 7.0151 0.0000 2007.01.04 12:22 7.0340 0.00 0.00 -196.91 1 350.58
9608196 2007.01.03 21:27 buy 5.00 usdsgd 1.5350 1.5320 0.0000 2007.01.04 14:50 1.5370 0.00 0.00 58.68 650.62
9608215 2007.01.03 21:29 buy 5.00 usddkk 5.6635 5.6767 0.0000 2007.01.04 12:22 5.6863 0.00 0.00 43.54 2 004.82
9608244 2007.01.03 21:30 buy 5.00 usdsek 6.8640 6.8906 0.0000 2007.01.04 12:22 6.8998 0.00 0.00 0.00 67.67 2 594.28
9609733 2007.01.03 23:09 sell 5.00 eurusd 1.3168 1.3115 0.0000 2007.01.04 12:22 1.3105 0.00 0.00 52.50 3 150.00
9609734 2007.01.03 23:09 sell 5.00 gbpusd 1.9510 1.9506 0.0000 2007.01.04 07:48 1.9506 0.00 0.00 -21.00 140.00
9609736 2007.01.03 23:09 sell 5.00 gbpjpy 232.81 232.11 0.00 2007.01.04 12:22 232.05 0.00 0.00 -292.71 2 228.55
9616006 2007.01.04 07:56 sell 5.00 gbpusd 1.9507 1.9446 0.0000 2007.01.04 12:22 1.9443 0.00 0.00 0.00 2 240.00
9625809 2007.01.04 12:36 buy 5.00 usdzar 7.0507 7.0655 0.0000 2007.01.04 14:54 7.0740 0.00 0.00 0.00 1 646.88
9625914 2007.01.04 12:43 buy 5.00 usdsek 6.9085 6.9172 0.0000 2007.01.04 14:50 6.9222 0.00 0.00 0.00 989.57
9625930 2007.01.04 12:44 sell 5.00 gbpjpy 231.90 231.75 0.00 2007.01.04 13:31 231.75 0.00 0.00 0.00 440.62
9625982 2007.01.04 12:47 buy 5.00 usdjpy 119.36 118.37 0.00 2007.01.04 15:42 119.37 0.00 0.00 0.00 41.89
9626003 2007.01.04 12:49 buy 5.00 eurchf 1.6137 1.6089 0.0000 2007.01.04 13:51 1.6139 0.00 0.00 0.00 81.21
9626019 2007.01.04 12:50 sell 5.00 eurusd 1.3097 1.3194 0.0000 2007.01.04 15:38 1.3093 0.00 0.00 0.00 200.00
9626022 2007.01.04 12:50 sell 5.00 gbpusd 1.9436 1.9454 0.0000 2007.01.04 21:40 1.9433 0.00 0.00 0.00 105.00
9626041 2007.01.04 12:52 sell 5.00 chfjpy 96.86 96.84 96.39 2007.01.04 13:57 96.84 0.00 0.00 0.00 83.90
9627627 2007.01.04 13:55 buy 5.00 usdzar 7.0452 7.0660 0.0000 2007.01.04 14:54 7.0732 0.00 0.00 0.00 1 979.30
9627737 2007.01.04 13:58 sell 5.00 chfjpy 96.76 96.75 0.00 2007.01.04 20:58 96.64 0.00 0.00 0.00 503.82
9627947 2007.01.04 14:03 sell 5.00 gbpusd 1.9450 1.9490 0.0000 2007.01.04 14:24 1.9441 0.00 0.00 0.00 315.00
9628629 2007.01.04 14:24 sell 5.00 gbpjpy 231.61 232.55 0.00 2007.01.04 20:58 231.56 0.00 0.00 0.00 146.94
9629742 2007.01.04 14:45 buy 5.00 usdnok 6.3212 6.3122 0.00 2007.01.04 15:45 6.3221 0.00 0.00 0.00 71.18
9630224 2007.01.04 14:56 buy 5.00 usdzar 7.0925 7.0840 0.0000 2007.01.04 15:58 7.0940 0.00 0.00 0.00 105.72
9630229 2007.01.04 14:56 buy 5.00 usdzar 7.0937 7.0855 0.0000 2007.01.04 20:59 7.0955 0.00 0.00 0.00 126.84
9630237 2007.01.04 14:56 buy 5.00 usdzar 7.0925 7.0842 0.0000 2007.01.04 15:58 7.0940 0.00 0.00 0.00 105.72
9630239 2007.01.04 14:56 buy 5.00 usdzar 7.0925 7.0840 0.0000 2007.01.04 15:59 7.0930 0.00 0.00 0.00 35.25
9633155 2007.01.04 15:54 sell 5.00 eurusd 1.3086 1.3184 0.0000 2007.01.04 17:19 1.3084 0.00 0.00 0.00 100.00
9633872 2007.01.04 16:11 sell 5.00 gbpusd 1.9466 1.9454 0.0000 2007.01.04 16:30 1.9451 0.00 0.00 0.00 525.00
9633889 2007.01.04 16:11 sell 5.00 eurusd 1.3110 1.3102 0.0000 2007.01.04 16:30 1.3095 0.00 0.00 0.00 750.00
9633992 2007.01.04 16:12 sell 5.00 gbpusd 1.9467 1.9454 0.0000 2007.01.04 16:29 1.9451 0.00 0.00 0.00 560.00
9634984 2007.01.04 16:37 sell 5.00 gbpusd 1.9443 1.9443 0.0000 2007.01.04 17:18 1.9439 0.00 0.00 0.00 140.00
9634988 2007.01.04 16:37 sell 5.00 gbpusd 1.9442 1.9541 0.0000 2007.01.04 17:19 1.9441 0.00 0.00 0.00 35.00
9634989 2007.01.04 16:37 sell 5.00 gbpusd 1.9442 1.9442 0.00 2007.01.04 17:18 1.9441 0.00 0.00 0.00 35.00
9634990 2007.01.04 16:37 sell 5.00 gbpusd 1.9442 1.9442 0.00 2007.01.04 17:18 1.9441 0.00 0.00 0.00 35.00
9636610 2007.01.04 17:21 sell 5.00 gbpusd 1.9444 1.9456 0.0000 2007.01.04 21:37 1.9437 0.00 0.00 0.00 245.00
9636613 2007.01.04 17:21 sell 5.00 gbpusd 1.9444 1.9456 0.0000 2007.01.04 21:37 1.9437 0.00 0.00 0.00 245.00
9636617 2007.01.04 17:21 sell 5.00 gbpusd 1.9443 1.9456 0.0000 2007.01.04 21:37 1.9437 0.00 0.00 0.00 210.00
9636619 2007.01.04 17:21 sell 5.00 gbpusd 1.9443 1.9456 0.0000 2007.01.04 21:37 1.9437 0.00 0.00 0.00 210.00
9636622 2007.01.04 17:21 sell 5.00 gbpusd 1.9443 1.9456 0.0000 2007.01.04 21:37 1.9436 0.00 0.00 0.00 245.00
9640117 2007.01.04 20:34 sell 5.00 eurusd 1.3083 1.3102 0.0000 2007.01.04 21:38 1.3080 0.00 0.00 0.00 150.00
0.00 0.00 -938.28 64 148.69
Closed P/L: 63 210.41
Open Trades:
Ticket Open Time Type Lots Item Price S/L T/P Price Commission Taxes Swap Profit
No transactions
0.00 0.00 0.00 0.00
Floating P/L: 0.00
Working Orders:
Ticket Open Time Type Lots Item Price S / L T / P Market Price
No transactions

Summary:
Deposit/Withdrawal: 5 000.00 Credit Facility: 0.00
Closed Trade P/L: 63,210.41 Floating P/L: 0.00 Margin: 0.00
Balance: 68 210.41 Equity: 68 210.41 Free Margin: 68 210.41
 
<br/ translate="no"> solandr thanks for the recommendation :)))
I tried for a month not to limit myself to one currency pair...
Do you manage to track 19 currency pairs manually?
or do you use EA on all pairs simultaneously?
Share the secret:)))

Of course I use EA. The EA is the same, it works with exactly the same algorithm on all currency pairs, purely based on graphical drawings. Just for each currency pair, the expert advisor uses different magiknumber for orders management. The idea behind the Expert Advisor is to identify the turning points by the first and second order convergent regressions described above in this thread. Further, as soon as the price leaves the 96% confidence interval for both regressions of the first and second orders, we consider ourselves to be at a turning point. The calculation is performed on the chart D1. Then on the period of M30 we look for signs of a reversal for entering a position on trend reversal. These signs may be, for example, a break in the downward or upward channel of a linear regression, coinciding with the trend that has occurred recently. Or the maximum or the minimum of the previous day can be used as evidence of the reversal. I have not yet decided which of these 2 methods is more reliable. When breaking the trend on M30 we have on average 2x shorter stops than when breaking the high/low of the previous day. But the probability of getting a stop loss when entering a position when breaking the linear regression channel is about 2 times higher than when entering a position when breaking the previous day's extremum. So I sit here experimenting. Rather, I observe how the Expert Advisor handles these entries. Not too much manual intervention is needed. Besides, I think I will be able to program it in the nearest future. The aim of the strategy is position trading with long holding and topping up. I think this is just quieter for me personally as there is always enough time to think. Here is an experimental example of orders (the orders were opened during modification of the Expert Advisor's algorithm and would have been better opened in the final variant, but I think the sense of position trading is clear from the report below).

Judging by this data on this symbol I have made 511 points of profit within the period of 21.12.2006-04.01.2007

. However, I cannot boast of any good result in general. I will continue my experiments.

PS: By the way, congratulations to all participants with reaching 100th page!!! :o)
 
Folks, I want to ask a question as we go along. I wanted to play around with the Hearst index, under the impression of this thread. But I absolutely do not understand how to calculate it! Or rather, how to calculate the entire value for a series is clear and very well described. But we are interested in the values in points! What is the Hurst index in a point - it is absolutely unclear. Say, we can count in overlapping windows. But with a small window we cannot understand what we will get as there is no point in talking about any statistics. In case of a large window, Hirst will be too averaged and generally will not be of much interest. Any advice on how to get out of this situation ?

Oops... Sorry for being silly. The solution is this. We look at Hearst on the daily chart and calculate it on the minute chart. Am I right ?
1...192193194195196197198199200201202203204205206...309
New comment
Reason: