<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:sy="http://purl.org/rss/1.0/modules/syndication/" > <channel><title>Comments on: Tick data &#8211; 99% modeling quality with Metatrader 4</title> <atom:link href="http://eareview.net/tick-data/feed" rel="self" type="application/rss+xml" /><link>http://eareview.net</link> <description>Metatrader expert advisors dissected.</description> <lastBuildDate>Thu, 17 May 2012 09:19:15 +0000</lastBuildDate> <sy:updatePeriod>hourly</sy:updatePeriod> <sy:updateFrequency>1</sy:updateFrequency> <generator>http://wordpress.org/?v=3.3.2</generator> <item><title>By: birt</title><link>http://eareview.net/tick-data/comment-page-18#comment-11168</link> <dc:creator>birt</dc:creator> <pubDate>Wed, 16 May 2012 12:46:28 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11168</guid> <description>It is using the fixed spread saved in the file. Using fixed spread with an FXT built with real spread is a bad idea, you should build another FXT instead.</description> <content:encoded><![CDATA[<p>It is using the fixed spread saved in the file. Using fixed spread with an FXT built with real spread is a bad idea, you should build another FXT instead.</p> ]]></content:encoded> </item> <item><title>By: Adam</title><link>http://eareview.net/tick-data/comment-page-18#comment-11167</link> <dc:creator>Adam</dc:creator> <pubDate>Wed, 16 May 2012 10:36:04 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11167</guid> <description>Gday Birt,I&#039;ve downloaded and successfuly run 99% variable spread testing, using your script to choose true (use variable spread). When I uploaded data I inserted 0.6 commission, no problem.All works very well, thank you for the instruction.My question is if I set set the use variable spread script to : false, what spread does it use ?I assume setting it to false does fix  the spread, and the tests shows a considerable better result, ie much easier to pass, NOT that I am looking for an easier test to pass. I am looking to compare a varaible spread test versus fixed spread.Thanks, your insights/advice is greatly valued.Cheers, Adam</description> <content:encoded><![CDATA[<p>Gday Birt,</p><p>I&#8217;ve downloaded and successfuly run 99% variable spread testing, using your script to choose true (use variable spread). When I uploaded data I inserted 0.6 commission, no problem.</p><p>All works very well, thank you for the instruction.</p><p>My question is if I set set the use variable spread script to : false, what spread does it use ?</p><p>I assume setting it to false does fix  the spread, and the tests shows a considerable better result, ie much easier to pass, NOT that I am looking for an easier test to pass. I am looking to compare a varaible spread test versus fixed spread.</p><p>Thanks, your insights/advice is greatly valued.</p><p>Cheers, Adam</p> ]]></content:encoded> </item> <item><title>By: birt</title><link>http://eareview.net/tick-data/comment-page-18#comment-11115</link> <dc:creator>birt</dc:creator> <pubDate>Thu, 10 May 2012 10:52:03 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11115</guid> <description>You also have to follow the rest of the instructions at http://eareview.net/tick-data/convert-tick-data and http://eareview.net/tick-data/tick-data-backtesting... You don&#039;t have to import anything in the history center.</description> <content:encoded><![CDATA[<p>You also have to follow the rest of the instructions at <a href="http://eareview.net/tick-data/convert-tick-data" rel="nofollow">http://eareview.net/tick-data/convert-tick-data</a> and <a href="http://eareview.net/tick-data/tick-data-backtesting" rel="nofollow">http://eareview.net/tick-data/tick-data-backtesting</a>&#8230; You don&#8217;t have to import anything in the history center.</p> ]]></content:encoded> </item> <item><title>By: Leo</title><link>http://eareview.net/tick-data/comment-page-18#comment-11114</link> <dc:creator>Leo</dc:creator> <pubDate>Thu, 10 May 2012 10:11:41 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11114</guid> <description>Hi Birt, I have downloaded 1 month of data from Dukascopy website, then followed your instructions up till converting csv to fxt. All files are on the right folders. Could i then open my mt4 history center and import the 1 month data ? is it the fxt or csv file that i import ? Thank you so much for your great details in explaining everything....Cheers...</description> <content:encoded><![CDATA[<p>Hi Birt,<br /> I have downloaded 1 month of data from Dukascopy website, then followed your instructions up till converting csv to fxt. All files are on the right folders.<br /> Could i then open my mt4 history center and import the 1 month data ? is it the fxt or csv file that i import ?<br /> Thank you so much for your great details in explaining everything&#8230;.</p><p>Cheers&#8230;</p> ]]></content:encoded> </item> <item><title>By: birt</title><link>http://eareview.net/tick-data/comment-page-18#comment-11017</link> <dc:creator>birt</dc:creator> <pubDate>Sun, 29 Apr 2012 23:07:13 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11017</guid> <description>Like I told you before, the &lt;i&gt;Birt&#039;s patch&lt;/i&gt; script is no longer supported. There must be a problem with your HST files but I couldn&#039;t say what exactly.</description> <content:encoded><![CDATA[<p>Like I told you before, the <i>Birt&#8217;s patch</i> script is no longer supported. There must be a problem with your HST files but I couldn&#8217;t say what exactly.</p> ]]></content:encoded> </item> <item><title>By: George</title><link>http://eareview.net/tick-data/comment-page-18#comment-11016</link> <dc:creator>George</dc:creator> <pubDate>Sun, 29 Apr 2012 23:03:54 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11016</guid> <description>Hi Birt,Well, I am getting the same problem again  :( It&#039;s weird.... Yesterday was working fine after repeating the CSv2FXT process for M30 and M1 TF.... Today I am getting 0 as the values for Ema&#039;s indicators...so is not calculating values properly... The code is Ok. Any clue ? sorry to bother you about this...Thanks !!George</description> <content:encoded><![CDATA[<p>Hi Birt,</p><p>Well, I am getting the same problem again <img src='http://cdn8.eareview.net/wp-includes/images/smilies/icon_sad.gif' alt=':(' class='wp-smiley' /><br /> It&#8217;s weird&#8230;.<br /> Yesterday was working fine after repeating the CSv2FXT process for M30 and M1 TF&#8230;.<br /> Today I am getting 0 as the values for Ema&#8217;s indicators&#8230;so is not calculating values properly&#8230;<br /> The code is Ok.<br /> Any clue ? sorry to bother you about this&#8230;</p><p>Thanks !!</p><p>George</p> ]]></content:encoded> </item> <item><title>By: Pirako</title><link>http://eareview.net/tick-data/comment-page-18#comment-11001</link> <dc:creator>Pirako</dc:creator> <pubDate>Sat, 28 Apr 2012 07:59:56 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-11001</guid> <description>Found it myself- thanks! Since i got &quot;serious&quot; in EA programming i might switch to the walk forward analyzer soon! :)</description> <content:encoded><![CDATA[<p>Found it myself- thanks! Since i got &#8220;serious&#8221; in EA programming i might switch to the walk forward analyzer soon! <img src='http://cdn8.eareview.net/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /></p> ]]></content:encoded> </item> <item><title>By: George</title><link>http://eareview.net/tick-data/comment-page-18#comment-10999</link> <dc:creator>George</dc:creator> <pubDate>Fri, 27 Apr 2012 21:38:32 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-10999</guid> <description>Hi Birt,I repeated the CSV2FXT process for 1M and M30 timeframes and now is working fine. The code was ok. Before I did it with the option to create multiple FXT files in a single run. So something went wrong with that process... I guess. Just to give you a feedback.Thanks a lot.</description> <content:encoded><![CDATA[<p>Hi Birt,</p><p>I repeated the CSV2FXT process for 1M and M30 timeframes and now is working fine. The code was ok.<br /> Before I did it with the option to create multiple FXT files in a single run.<br /> So something went wrong with that process&#8230; I guess.<br /> Just to give you a feedback.</p><p>Thanks a lot.</p> ]]></content:encoded> </item> <item><title>By: George</title><link>http://eareview.net/tick-data/comment-page-18#comment-10995</link> <dc:creator>George</dc:creator> <pubDate>Fri, 27 Apr 2012 13:52:13 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-10995</guid> <description>Ok Birt,Surely I have something wrong with my installation or my code. I will check. Thanks for your time and your answer. George</description> <content:encoded><![CDATA[<p>Ok Birt,</p><p>Surely I have something wrong with my installation or my code. I will check.<br /> Thanks for your time and your answer.<br /> George</p> ]]></content:encoded> </item> <item><title>By: Pirako</title><link>http://eareview.net/tick-data/comment-page-18#comment-10993</link> <dc:creator>Pirako</dc:creator> <pubDate>Fri, 27 Apr 2012 10:21:31 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-10993</guid> <description>Hey birt :) ive been using your patch a lot and it worked just fine! (i dont mind the crashes when closing the tester.. anyways i thought might it be possible to have another patch which rlses the memory alteration to prevent from crashing?)But my actual question: Do you still have an installer for the old MT4 version compatible with the patch? I messed around with my installation and end up with a fresh install to 413 :( so the patch doesnt work anymore as stated out.Would appreciate a lil help :)@ geroge:i do multi TF backtesting with different indicators all the time.. no problems there :) check ur code!greetings</description> <content:encoded><![CDATA[<p>Hey birt <img src='http://cdn8.eareview.net/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /><br /> ive been using your patch a lot and it worked just fine! (i dont mind the crashes when closing the tester.. anyways i thought might it be possible to have another patch which rlses the memory alteration to prevent from crashing?)</p><p>But my actual question: Do you still have an installer for the old MT4 version compatible with the patch? I messed around with my installation and end up with a fresh install to 413 <img src='http://cdn8.eareview.net/wp-includes/images/smilies/icon_sad.gif' alt=':(' class='wp-smiley' /> so the patch doesnt work anymore as stated out.</p><p>Would appreciate a lil help <img src='http://cdn8.eareview.net/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /></p><p>@ geroge:</p><p>i do multi TF backtesting with different indicators all the time.. no problems there <img src='http://cdn8.eareview.net/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> check ur code!</p><p>greetings</p> ]]></content:encoded> </item> <item><title>By: birt</title><link>http://eareview.net/tick-data/comment-page-18#comment-10992</link> <dc:creator>birt</dc:creator> <pubDate>Fri, 27 Apr 2012 08:15:03 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-10992</guid> <description>First of all, please note that I am not offering any support for the &lt;i&gt;Birt&#039;s patch&lt;/i&gt; script.These being said, I suggest doing some logging to check what&#039;s wrong in your particular case because I&#039;m 100% certain that EAs using indicators on multiple timeframes work fine in tick data backtesting. In fact, I&#039;ve checked it yesterday for a TDS customer and if you don&#039;t want to take my word for it, write the following in an EA and backtest it on two different timeframes on ~1 month of data:&lt;code&gt; int start() { if (iBars(Symbol(), PERIOD_M1) &lt; 30) return (0); static datetime lastM5bar; if (lastM5bar != iTime(Symbol(), PERIOD_M5, 0)) { lastM5bar = iTime(Symbol(), PERIOD_M5, 0); double ima = iMA(Symbol(), PERIOD_M1, 30, 0, MODE_EMA, PRICE_WEIGHTED, 0); Print(DoubleToStr(ima, 5)); } return(0); } &lt;/code&gt;If you did everything right the results printed in the log should be identical. If they aren&#039;t, make sure that: &lt;ul&gt; &lt;li&gt;The same CSV was used for generating the FXT/HST for both timeframes.&lt;/li&gt; &lt;li&gt;The same GMT/DST settings were used.&lt;/li&gt; &lt;li&gt;The files were copied to the correct location.&lt;/li&gt; &lt;/ul&gt;</description> <content:encoded><![CDATA[<p>First of all, please note that I am not offering any support for the <i>Birt&#8217;s patch</i> script.</p><p>These being said, I suggest doing some logging to check what&#8217;s wrong in your particular case because I&#8217;m 100% certain that EAs using indicators on multiple timeframes work fine in tick data backtesting. In fact, I&#8217;ve checked it yesterday for a TDS customer and if you don&#8217;t want to take my word for it, write the following in an EA and backtest it on two different timeframes on ~1 month of data:</p><p><code><br /> int start()<br /> {<br /> if (iBars(Symbol(), PERIOD_M1) < 30) return (0);<br /> static datetime lastM5bar;<br /> if (lastM5bar != iTime(Symbol(), PERIOD_M5, 0)) {<br /> lastM5bar = iTime(Symbol(), PERIOD_M5, 0);<br /> double ima = iMA(Symbol(), PERIOD_M1, 30, 0, MODE_EMA, PRICE_WEIGHTED, 0);<br /> Print(DoubleToStr(ima, 5));<br /> }<br /> return(0);<br /> }<br /> </code></p><p>If you did everything right the results printed in the log should be identical. If they aren't, make sure that:</p><ul><li>The same CSV was used for generating the FXT/HST for both timeframes.</li><li>The same GMT/DST settings were used.</li><li>The files were copied to the correct location.</li></ul><p></code></p> ]]></content:encoded> </item> <item><title>By: george</title><link>http://eareview.net/tick-data/comment-page-18#comment-10987</link> <dc:creator>george</dc:creator> <pubDate>Thu, 26 Apr 2012 20:06:21 +0000</pubDate> <guid isPermaLink="false">http://eareview.net/?page_id=19#comment-10987</guid> <description>Hi Birt,Thanks for your answer !My EA works with 2 ema&#039;s : one on the current timeframe (30M,1H) and the other ema with 1 Minute period. I run the EA in a MT4 normal strategy tester under 30M timeframe and it works fine. This means that the signal based on the 30M and 1 minute ema opens the orders correctly.When I used the MT4 version with birt&#039;s patch, it seems that the EA does not recognize the 1M ema indicator, since no orders are opened. If I delete the portion of the code corresponding to the 1M ema, the EA does open the orders. So we may conclude that the 1M period ema indicator is not recognized for any reason. If this is happening with other EA&#039;s then this could cause misleading results.I am using Oanda v. 401, dukascopy Jforex data and Birt&#039;s patch version for r. 401. All HST files for each period are in the history/Oanda practice folder.This is the portion of the code I have :***** Current timefarme 30M EMA&#039;s  : SlowMAbar0 = iMA(NULL,0,SlowMAPeriod,SlowMAShift,SlowMAMethod,SlowMAAplliedPrice,MASignalBar); SlowMAbar1 = iMA(NULL,0,SlowMAPeriod,SlowMAShift,SlowMAMethod,SlowMAAplliedPrice,MASignalBar+1); SlowMAbar2 = iMA(NULL,0,SlowMAPeriod,SlowMAShift,SlowMAMethod,SlowMAAplliedPrice,MASignalBar+2); FastMAbar0 = iMA(NULL,0,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar); FastMAbar1 = iMA(NULL,0,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar+1); FastMAbar2 = iMA(NULL,0,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar+2); RSIbar0 = iRSI(NULL,0,RSIPeriod,RSIAppliedPrice,RSISignalBar); RSIbar1 = iRSI(NULL,0,RSIPeriod,RSIAppliedPrice,RSISignalBar+1);****** 1 minute EMA: MSlowMAbar0 = iMA(NULL,1,SlowMAPeriod,SlowMAShift,0,SlowMAAplliedPrice,MASignalBar); MSlowMAbar1 = iMA(NULL,1,SlowMAPeriod,SlowMAShift,0,SlowMAAplliedPrice,MASignalBar+1); MFastMAbar0 = iMA(NULL,1,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar); MFastMAbar1 = iMA(NULL,1,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar+1);if ((FastMAbar1  SlowMAbar0) &#124;&#124; (FastMAbar2  SlowMAbar0)) //&amp;&amp; RSIbar1 = RSISignalLevel) { if (TradeNotPlacedYet()) { if (MFastMAbar1  MSlowMAbar0) { OpenBuyOrder();  ****** NOT OPENING ORDER HERE, IF I DELETE THE if (MFastM... portion it does open the order correctly... } } } if ((FastMAbar1 &gt; SlowMAbar1 &amp;&amp; FastMAbar0  SlowMAbar2 &amp;&amp; FastMAbar0  RSISignalLevel &amp;&amp; RSIbar0  MSlowMAbar1 &amp;&amp; MFastMAbar0 &lt; MSlowMAbar0) { OpenSellOrder(); } } }</description> <content:encoded><![CDATA[<p>Hi Birt,</p><p>Thanks for your answer !</p><p>My EA works with 2 ema&#8217;s : one on the current timeframe (30M,1H) and the other ema with 1 Minute period. I run the EA in a MT4 normal strategy tester under 30M timeframe and it works fine. This means that the signal based on the 30M and 1 minute ema opens the orders correctly.</p><p>When I used the MT4 version with birt&#8217;s patch, it seems that the EA does not recognize the 1M ema indicator, since no orders are opened. If I delete the portion of the code corresponding to the 1M ema, the EA does open the orders. So we may conclude that the 1M period ema indicator is not recognized for any reason. If this is happening with other EA&#8217;s then this could cause misleading results.</p><p>I am using Oanda v. 401, dukascopy Jforex data and Birt&#8217;s patch version for r. 401. All HST files for each period are in the history/Oanda practice folder.</p><p>This is the portion of the code I have :</p><p>***** Current timefarme 30M EMA&#8217;s  :<br /> SlowMAbar0 = iMA(NULL,0,SlowMAPeriod,SlowMAShift,SlowMAMethod,SlowMAAplliedPrice,MASignalBar);<br /> SlowMAbar1 = iMA(NULL,0,SlowMAPeriod,SlowMAShift,SlowMAMethod,SlowMAAplliedPrice,MASignalBar+1);<br /> SlowMAbar2 = iMA(NULL,0,SlowMAPeriod,SlowMAShift,SlowMAMethod,SlowMAAplliedPrice,MASignalBar+2);</p><p> FastMAbar0 = iMA(NULL,0,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar);<br /> FastMAbar1 = iMA(NULL,0,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar+1);<br /> FastMAbar2 = iMA(NULL,0,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar+2);</p><p> RSIbar0 = iRSI(NULL,0,RSIPeriod,RSIAppliedPrice,RSISignalBar);<br /> RSIbar1 = iRSI(NULL,0,RSIPeriod,RSIAppliedPrice,RSISignalBar+1);</p><p>****** 1 minute EMA:<br /> MSlowMAbar0 = iMA(NULL,1,SlowMAPeriod,SlowMAShift,0,SlowMAAplliedPrice,MASignalBar);<br /> MSlowMAbar1 = iMA(NULL,1,SlowMAPeriod,SlowMAShift,0,SlowMAAplliedPrice,MASignalBar+1);<br /> MFastMAbar0 = iMA(NULL,1,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar);<br /> MFastMAbar1 = iMA(NULL,1,FastMAPeriod,FastMAShift,FastMAMethod,FastMAAplliedPrice,MASignalBar+1);</p><p> if ((FastMAbar1  SlowMAbar0) || (FastMAbar2  SlowMAbar0)) //&amp;&amp; RSIbar1 = RSISignalLevel)<br /> {<br /> if (TradeNotPlacedYet())<br /> {<br /> if (MFastMAbar1  MSlowMAbar0)<br /> {<br /> OpenBuyOrder();  ****** NOT OPENING ORDER HERE, IF I DELETE THE if (MFastM&#8230; portion it does open the order correctly&#8230;<br /> }<br /> }<br /> }<br /> if ((FastMAbar1 &gt; SlowMAbar1 &amp;&amp; FastMAbar0  SlowMAbar2 &amp;&amp; FastMAbar0  RSISignalLevel &amp;&amp; RSIbar0  MSlowMAbar1 &amp;&amp; MFastMAbar0 &lt; MSlowMAbar0)<br /> {<br /> OpenSellOrder();<br /> }<br /> }<br /> }</p> ]]></content:encoded> </item> </channel> </rss>
<!-- Performance optimized by W3 Total Cache. Learn more: http://www.w3-edge.com/wordpress-plugins/

Minified using disk: basic
Page Caching using disk: enhanced
Database Caching 12/28 queries in 0.037 seconds using memcached
Object Caching 619/636 objects using memcached
Content Delivery Network via cdn5.eareview.net

Served from: eareview.net @ 2012-05-17 12:59:05 -->
