|
/ | Forum |
  |
Ruptor
2009.06.30 18:16
Hi Please make any stored settings consistent in MT5 otherwise don't bother saving them. For instance I have 2 brokers and if I edit an EA in one broker then close down that broker and edit a file in the other broker MetaEditor still holds the file paths of the first broker. Many times I have done "save as" and saved the EA in the wrong broker directory because I hadn't realised the save directory stored location was the first broker instread of the second broker where the file was opened. |
 |
skatt
2009.06.30 23:56
There needs to be a way to test how EA handles balance deposits and withdrawals on demo accounts.
|
 |
TheEconomist
2009.07.01 10:17
Well well, today is July the 1st and counting down days... Guess it will be released on my birthday... July 20, when support for older MT4 builds gets suspended.
|
|
Jellybean
2009.07.04 13:21
For me, the lack of separate positions in one currency is just going back to the way I used to do it. No big deal. However, I'd really like to know what the situation is with stop-loss and take profit orders. I've read that they aren't allowed under the NFA ruling. Other platforms are offering OCOs (one cancels other) as an alternative to stop-loss and take profit. Is this what is planned for MT5? If so, will it be possible with MT5 to have automatic/pending follow-on or contingent orders? For example, enter with a pending Buy Stop; if that executes, then automatically place an OCO Stop & Limit order? Cheers Jellybean |
|
bredin
2009.07.04 16:30
There has been a claim that a hedge position makes no difference in the long run, compared to single position but something did occur to me. follow this: with hedge open long going for 30 pips profit. (pay 2 pips spread) price moves against so scalp 5 pips short. (pay 2 pips spread) with hedge, pay spread twice, +35 pips net profit without hedge, open long close long, open short close short open long three orders so pay spread three times, not twice as in first example, 33 pips net profit. I may be wrong, but i cant see brokers charging the spread less times rather than more, if possible |
|
Jellybean
2009.07.05 04:18
bredin wrote >>
There has been a claim that a hedge position makes no difference in the long run, compared to single position but something did occur to me. follow this: with hedge open long going for 30 pips profit. (pay 2 pips spread) price moves against so scalp 5 pips short. (pay 2 pips spread) with hedge, pay spread twice, +35 pips net profit without hedge, open long close long, open short close short open long three orders so pay spread three times, not twice as in first example, 33 pips net profit. I may be wrong, but i cant see brokers charging the spread less times rather than more, if possible Hi Bredin I'm one of the ones that thinks there's no difference between separate positions or one total position per currency, except in how the trader has to manage their trading. I thought about your example because I also think examples are good to clarify things that are confusing. Here's a table with the same two examples you gave, but I've added example trade prices. I've assumed the trader entered at 1.0000, hedged or sold at 0.9995, exited the hedge or bought at 0.9990 and finally exited at 1.0030. I also account for the spread on the buys, because the you lose only one spread for each buy/sell pair of orders.
I still think the resulting profit and costs are the same, i.e. only two spreads paid in both cases. I do accept, though, that it's harder to manage short-term trading and longer-term trading on the same currency because yo don't have visibility of separately. Cheers Jellybean |
|
bredin
2009.07.05 04:56
Jellybean wrote >>
Hi Bredin I'm one of the ones that thinks there's no difference between separate positions or one total position per currency, except in how the trader has to manage their trading. I thought about your example because I also think examples are good to clarify things that are confusing. Here's a table with the same two examples you gave, but I've added example trade prices. I've assumed the trader entered at 1.0000, hedged or sold at 0.9995, exited the hedge or bought at 0.9990 and finally exited at 1.0030. I also account for the spread on the buys.
I still think the resulting profit and costs are the same, i.e. only two spreads paid in both cases. I do accept, though, that it's harder to manage short-term trading from longer-term trading on the same currency. Cheers Jellybean Now I see it, the short 'scalp' is actually no position and the scalped pips are in the second buy. Thats going to take some getting used to :) |
|
apparelink
2009.07.13 04:49
TheEconomist wrote >>
Even if the broker reconstructs options as CFDs to offer them to the trader, all possibilities of trading should be left open. I'd love an MT5 that could be like that. Hi Economist, I completely agree with you. This is why I am always on the look out for MT4 brokers who can provide other equity, commodity and index symbols. Unfortunately most MT4 brokers do NOT offer them. Such lack of crucial intermarket information is forcing a lot of FX traders to fly blind and lose money quickly. If I had to choose, I'd prefer a broker who can offer more symbols in their feed rather than the one with the tightest spreads. I understand it costs more for the broker to provide more symbols. That's why many experienced traders also use a second platform to help them with their analysis (e.g., eSignal and TradeStation) because they are not anemic in historical data for other assets. This is not a short coming of the MT4 platform or MetaQuotes but rather the brokers. The only one MT4 broker I have found so far that offers all of these symbols you are talking about is a Russia-based broker called BroCo. Their spreads are not the best and they don't provide fractional pips. Also, I found there is a higher number of UK-based MT4 brokers offering CFDs but very few US-based brokers do. I just finished a correlation indicator and it supports 8 different correlated pairs (a lot more robust than many other correlation indicators I can download from the Internet) and tested that on BroCo and it almost made me cry (tears of joy). Correlation analysis reveals fundamental driving forces in ways NO other indicators can do. It also cast doubts on many conventional wisdoms like gold price driving AUDUSD or NZDUSD because a positive correlation comes and goes and it is DANGEROUS for many to hold this belief unquestioned without looking at how it changes over time. For example, AUDUSD correlation with gold actually DROPPED OUT in the last few months and only came back in the last couple of weeks. The following screenshot confirms the current U.S. dollar weakness. Scott |
|
TheEconomist
2009.07.13 09:57
apparelink wrote >>
Hi Economist, I completely agree with you. This is why I am always on the look out for MT4 brokers who can provide other equity, commodity and index symbols. Unfortunately most MT4 brokers do NOT offer them. Such lack of crucial intermarket information is forcing a lot of FX traders to fly blind and lose money quickly. If I had to choose, I'd prefer a broker who can offer more symbols in their feed rather than the one with the tightest spreads. I understand it costs more for the broker to provide more symbols. That's why many experienced traders also use a second platform to help them with their analysis (e.g., eSignal and TradeStation) because they are not anemic in historical data for other assets. This is not a short coming of the MT4 platform or MetaQuotes but rather the brokers. The only one MT4 broker I have found so far that offers all of these symbols you are talking about is a Russia-based broker called BroCo. Their spreads are not the best and they don't provide fractional pips. Also, I found there is a higher number of UK-based MT4 brokers offering CFDs but very few US-based brokers do. I just finished a correlation indicator and it supports 8 different correlated pairs (a lot more robust than many other correlation indicators I can download from the Internet) and tested that on BroCo and it almost made me cry (tears of joy). Correlation analysis reveals fundamental driving forces in ways NO other indicators can do. It also cast doubts on many conventional wisdoms like gold price driving AUDUSD or NZDUSD because a positive correlation comes and goes and it is DANGEROUS for many to hold this belief unquestioned without looking at how it changes over time. For example, AUDUSD correlation with gold actually DROPPED OUT in the last few months and only came back in the last couple of weeks. The following screenshot confirms the current U.S. dollar weakness. Scott Hi Scott, Well, I knew about Broco. I use their demo too. I could see it's not only the lack of crucial intermarket information (because there is that information, at least for the last months of fx), but rather the lack OF THE GODDAMN MULTIASSET BACKTESTER is what is forcing traders to lose money quickly, because correlation systems can't be fine tuned, so that's why they were dropped in favor of monoasset indicator-bet systems. Your indicator works well on AUDUSD, I see support and resistance plotted well (though there are mistakes around March 4 and April 29) but if an indicator would require trading on both analysed assets, most likely would be an off-chart indicator and need a multiasset backtester. I'm sure however that Broco might be one of the first brokers who would offer MT5, even with the options included in the offer. Retail trading might be so different after the launch of MT5, that many "fx traders" probably wouldn't even like to call themselves like that anymore, when they see how wide becomes trading when you can backtest and trade everything that gets in your mind. |
|
apparelink
2009.07.13 14:10
TheEconomist wrote >>
because correlation systems can't be fine tuned, so that's why they were dropped in favor of monoasset indicator-bet systems. Your indicator works well on AUDUSD, I see support and resistance plotted well (though there are mistakes around March 4 and April 29) but if an indicator would require trading on both analysed assets, most likely would be an off-chart indicator and need a multiasset backtester. Hi Economist, I don't quite understand why correlation systems cannot be fine-tuned. I remember from one of your posts that you have explored correlation in the past without much success. I would really love to hear your findings if you don't mind sharing. BTW, my CORREL indicator works for any currency pair, not just AUDUSD (see screenshots). There is a user option UseAutoPairSelection which doesn't require the user to manually select the appropriate symbols for correlation analysis as I have coded the indicator to work with different MT4 brokers (detected using AccountCompany()).
While most MT4 brokers use standard symbols there are brokers who do not follow the convention. Examples: What I am researching at the moment is how to use coefficients to provide reliable signals. Finding the causality is probably the most difficult and at best a guess. For example, most people would use hedging strategy with the knowledge gained from correlations. However, the CORREL index lines can shift (trend), relign (cross) and misalign (diverge) over time, much like a price MA line. Line crossing isn't of big interest to me unless it happens above the +/-0.75 region. Some suggest that if one pair breaks out, the other highly correlated pair is likely to break R1, R2, R3, etc., as well. This allows traders to take advantage of the potential time lag in order to predict the price action of other highly correlated currency pairs and crosses (this approach will only work if automated because a human trader cannot possibly monitor multiple pairs on the same screen to identify the best calculated opportunities and act fast enough). However when two symbols are too highly correlated (over 0.95), the price action of the second pair reflects that almost immediately and there isn't much room to exploit the time lag. Perhaps correlation would be more valuable for inter-equity, inter-commodity or inter-futurers market analysis rather than within the spot FX market as the time lag between intermarkets may be greater. My idea of backtesting is that, you don't need to test that in Strategy Tester (you cannot anyway); you only need to know the probability that the correlation indicator (using your interpretation and strategy) is reliable. If it more than 50% of the time, you already have the odds stack in your favor; the rest is about good money management to help you stay in the game. FYI, I included the following links which I researched about correlations: Concepts and theories |
