Agility TRADER - Automate your ideas and expand your mind

hello Dave, thanks a lot for your contribution, looks very interesting

do you think its possible to manage a buy grid and sell grid of orders with martingale with your ea?

​In relation to the uploaded Set file from Admir, you have set as your Buy Criteria as Price being less than Support (measured over the 100 bars) with a buffer of 5 pips for a Buy signal and Price being greater than Resistance (measured over the 100 bars) with a buffer of 5 pips for a Sell signal.

The point is that you really don't provide any Entry rules as all you are saying is Buy at any price below Support + 5 pips and Sell at any price Resistance - 5pips.

What you should be saying is -
Buy when the Close Price is less than Support + 5.
Sell when the Close Price is greater than Resistance - 5

See the example set file included in the zip file and please consider deleting your Post #38. (Post #39 refers)

​Sorry not possible.

Only one active buy or sell trade per chart.

Okay, yesterday had this problem too with nolagdot.
And with DivergenceRSI - cant find working params... maybe i do something wrong...
With support\resistance cant find good params (for example - open buy only if last 3 candles close grater then support (5-10 buffer) - will continue to experiment...

And... found some mistakes in .pdf manual
for example:
1) Pivot points ... what is Param1? timezone or pivot point?
2) Support... similarly... (in "Resistance" looks like no mistake)

and how to check for PSAR? Type1_psar (param1 = 0.02, param2 = 0.2) Equal Type2_param1 = 1 (2) ? No explain in PDF.
And can you add DMI (ADX param at least, for check trend. Type1_dmiadx (60_240_1440) last candle > type2_dmiadx 2 candles back... for example)?

Thanks, anyway you done great job.

​Can't see any problem in the documentation with Pivots. Timezone is Param1, Pivot Points is Param2, Buffer is Param3.

Support Param2 should be Buffer (thank you)

Support and Resistance need Param1 value to be at least 50 (ie. bars for calculation)

With PSAR - Use this type of analogy - Close crosses above (or below) PSAR. With Divergence/RSI, I would need to see your set file. (Upload as an attachment - rename to .txt)

no mistake in PDF about params MACD, STOCH? "Param1 represents the Fast MA, Param2 represents the Slow MA and Param3 represents the smoothing period."
Then i open test grafic, where STOCH indi show with param 3 10 3 too, and its wrong param ( good is 10/3/3, for example)

Great Work for community....Briliant...
Thanks Dave_C

Hello

May be a stupid question ..... i'm looking for a reliable source of history data (EURUSD 5mn.. ) any clue ?

http://www.forexfactory.com/showthread.php?p=2555813

Thanks kalawe... going to use duskcopy history web link

Has anyone tried creating a set file based off the 00 level trading thread? There are many successful traders using this method and would be great if we had a way to automate it using Agility Trader.

there is no RoundNumbers indi

Have no idea what you are saying. Both indicators are correct!

Hi Dave,
AWESOME work! Best tool out there.

Quick question, is it possible to add the LWMA as an option on the indicator choices?

Thanks again mate for a wonderful tool!


ST65

​Update for Agility Trader at Post #1.

Changes -

1. NonLagDot logic in EX4
2. change in companion program for XBarsAgo input BuyEntry Condition4
3. Amendments to documentation

For those of you who have been following my posts, you know I advocate back testing on only the most recent 3 months data.

RSI for me has always been one of the most reliable indicators. I've enclosed a set file (rename to .set) which is a good example for EUR 60mins

Weighted MA is the same as LWMA and is one of the choices.

Regards Dave

I've reproduced a post that I have made previously on another forum which I enclose for your consideration and comment.

QUOTE:
I wanted to touch on the issue of money management.

The EA has a range of techniques to control your investments in varying complexity

1. Fixed wager, wager as percentage of equity, wager as percentage of equity but based on the projected stop loss
2. Fixed stop loss and fixed take profit
3. Set stop loss and take profit on a nominated ATR (Average True Range)
4. Close trades on reversal of indicators, close trades on the TDI (traders dynamic index)
5. Trailing stop loss
6. Set stop loss to break even after a nominated pip profit.
7. Set take profit on a risk v's reward nominated ratio
8. Increase wager by a nominated multiple progressively on losing trades (up to a limit 20% of equity)
9. Increase wager if a nominated pip gain is not achieved (up to a limit 20% of equity)

For any given currency set and time period, the equity curve and maximum drawdown are progressively improved as points 1 to 9 are introduced.

We all are aware of the Martingdale system and how it has been comprehensively disproven. The casinos control this type of bet progression by setting minimum and maximum wagers. Typically the maximum is 100 times the minimum. For example a minimum $5 table will allow a single wager of $500. In Martingdale a progressive losing bet sequence would be 5, 10, 20, 40, 80, 160, 320 and then the final wager would be 500 but that bet would result in an overall loss even if it was a winning wager. So in reality a sequence of 7 losses only allowable to recover a profit of $5.

We have tools which are much more sophisticated, such as being able to regulate our risk v's reward ratio, minimum wager of .01 and a maximum wager of nominally around 50 so our minimum to maximum ratio is around 5000. This combined with the risk v's reward ratio and the ability to set our bet progression ratio to any figure give us a very powerful arsenal. Combine all this with our detection abilities (fundamental and technical), I sincerely believe we are only tapping the surface at the moment. Our most formidable opponents are the market makers and the major bankers who control the price flow. We all know how many EA's are in existence both commercially and through the forums which display great promise with curve fitting techniques and ultimately never hold up because of the constantly changing marketplace.

Last month we witnessed the great CHF debacle where many got burnt. Nobody really had the insight to see it coming but now we see some brokers actually advertising that they are going to write off client losses in a bid to retain customers.

This brings me to the point of this post, I have just completed a massive back testing exercise where all the money management factors were optimized on a continuous straddle strategy. Why this strategy - because we really never know how the market is going to behave so I am putting greater emphasis on money management. The most powerful weapon by far is the ability to leverage your investment.

In a nutshell, just say we have a potential investment of a percentage of our equity say 2% of $5000 ie. $100.

So we wager $100. Realistically what are our expectations, say win $100 or lose $100.

What if we viewed it like this - if we lowered our investment size but employed a trailing stop loss, set stop loss to breakeven, increased our wager size with consecutive losses, employed a risk reward ratio and set a minimum level for a result not to be treated as a loss.

Here is the basic result on the back test (AUDUSD 01112014 to 31012015 Cont Straddle 1 min chart)
Simple TP and SL with no trail $1930 and various increasing returns as each successive factor is introduced -
Up to the ... full optimization $28,429.

This is the astounding part - that return was produced from a initial investment of 0.5% of equity. The absolute essence of it is the stop loss multiplier of 2 (the result of an optimization from 1.1 to 2 with a step of .1) and the low initial wager. I have purposely capped the maximum wager at 20% of equity. (After that back to the .5%). The values achieved during back test will of course vary with scenarios. My preference is for a starting point of 1.5% of equity and a StopLossMultiplier of 1.2. This gives a progression of 16 losing trades before reaching the 20% cap of equity value. If you reduce the initial starting point percentage you are able to achieve a greater losing sequence before reaching 20% of equity. As a result of my back testing on all currencies and all time frames the 16 loss sequence seemed quite adequate.

The real point here is that you need to be in a position to make the most of that winning trade when it comes, and if it takes a number of break evens or low profit wins (because of the trailing stop loss caught up), you are giving yourself the best chance to capitalize. You have to accept that sometimes you will have a "bad hair day" when the trade simply goes pear shaped from the beginning. The majority of your trades will be in the break even or trailing stop loss scenario and this is where this money management strategy excels.

Do you get mismatched chart errors when backtesting. Here is a guide to clean up your data. The bad news is that you will have to delete any existing history files and reimport data

Data is available from http://www.forextester.com/data/datasources (skip the first line and column on data import options)

1. Shut down MT4
2. Turn off your internet (most important)
3. If you do not how to locate the hidden History directory, Restart MT4.
4. Delete the history data for the pairs that you intend to refresh. that is if you are cleaning EURUSD delete all the EURUSD.hst files. This directory can be located by Selecting File, Open Data Folder and selecting the History Directory.
5. If MT4 has been restarted to locate the MT4 directory, after you have deleted the .hst files, shut down MT4
6. Restart MT4
7. Select Tools, History Center and double click on the required currency pair and select 1 minute time frame. The table should be empty.
8. click on the import button.
9. select the datafile (csv extension) to be imported. If importing multiple files, start with the earliest data.
10. it may take 15 secs to load. If the file contents do not display in the table, click on Cancel and repeat the Tools etc. selection again.
11. select the correct GMT shift for your broker
12. click on the OK button
13. click on the Close button
14. select the 1 min chart for the currency pair being imported
15. Use the script PeriodConverter to convert the 1 minute data to 5,15,30,60,240 and 1440 minute time frames. You will have to rerun the script each time and will be asked to cancel the previous script each time.
16. by rechecking the History Center you will have a confirmation that the PeriodConverter has done its job.
17. Turn off MT4
18. Turn on the internet and restart MT4.
19. MT4 will automatically add its own data to the end of the imported data and you will have eliminated the errors and should achieve 90% accuracy with no mismatched chart errors.

What do you think about the idea of adding Quantum Indicator? I have had pretty good results running an EA based off this indicator.