Who uses Metastock?

Quote from intradaybill:

First of all you never said imbecile that the second code is for the last version only. I don't care about the last version. I talked about a problem I faced long ago. Second, check the first code because it has deceived you.. Some entries are right but some are not. You tried to force the backtester by applying a stop function but it does not produce the correct resutls for the exits as many of them are not at the open of next bar but at some arbitrary level.

Do not jump like a monkey up and down. This is a trading forum, not a zoo. You are dealing with someone who can nail you imbecile. As I said in the PM very politely to you, I did this a long time ago. You must have some vested interest to be jumping like this. You started inslulting and attacking. Everything is on record.

Your first code does not work well in version 5.40.

Example: On 03/22/2012 Entry on close: 32 - correct

Exit should be on open of 03/23/2012 at 32.10

Instead it exits at 31.98

You trying to emulate a simple system with complicated code is laughable to anyone who has coded systems and shows your panic mode.

Try again. Please no results form latest version when this was probably fixed. I was talking about a while ago. I made that clear.

Do not hide.

In regards to zoo you are behaving like a monkey for the last few posts and you don't realize it.
You are attacking me at the same level, hypocrite. Your sentences are just laughable.

Complicated code? What are you talking about again. Setoptions are just options set via AFL. The rest ist just 5/6 lines of standard code. It does not make any sense to talk to a child like I'm doing right now.

The only interest I had was that of a user who could be prepared for possible issues so I could mail them to support what you did not seem to have done yet.

I will test what you wrote on 5.4.

In addition we need someone neutral like Tim here because it's absolutely impossible to talk to someone like you.
 
There seems to be a problem with Yahoo's servers. The data download via downloader is different to the data downloaded manual via Yahoo's website.

For 03/22/2012 I have different prices for MSFT

So I will download the data manually from their site and re-test
 
Sorry Bill, but I don't know what you are doing. As for 03/22/2012 I'm getting correct entry and exit using 5.40

Here is a video (forget about the video quality it's just GIF format)
msft.gif


Since I'm not an asshole like others around us
here is the code seen in the video ready for copy and paste
( it's the same one as the first one from here http://www.elitetrader.com/vb/showthread.php?s=&postid=3569565#post3569565 )
Code:
Buy = 1;

Sell = Short = Cover = 0;
StopLevel = Param("N-bars", 1, 1, 100, 1 );
ApplyStop( stopTypeNBar, stopModeBars, StopLevel, 1, False, 0  );

BuyPrice = C; 
SellPrice = O;

SetOption("InitialEquity", 100000);
SetOption("FuturesMode", False );
SetOption("UsePrevBarEquityForPosSizing", True);
SetOption("ActivateStopsImmediately", False);
SetOption("Allowsamebarexit", True);
SetOption("AllowPositionShrinking", True);
SetOption("MaxOpenPositions", 1);
SetPositionSize( 1, spsShares );
//Reports
SetOption("PortfolioReportMode", 0);// 0-Trade list, 1- Detailed Log, 2- Summary, 3- No output
SetOption("GenerateReport", 1); // force generation of full report 

RoundLotSize = 1;

Here is the data I was using this time manually downloaded from Yahoo
download here http://finance.yahoo.com/q/hp?s=MSFT+Historical+Prices

Import it like this via the import wizard
choose
YMD,Open,High,Low,Close,Volume,Skip

and skip first 1 lines


I did use a fresh install of 5.40 in a virtual machine so I did not change any settings. And I'm getting same results in 5.50
 
Quote from funnyguy:

There seems to be a problem with Yahoo's servers. The data download via downloader is different to the data downloaded manual via Yahoo's website.

For 03/22/2012 I have different prices for MSFT

So I will download the data manually from their site and re-test

It's seems to be a Yahoo server issue . I have tried two independent Yahoo data downloaders both showing same corrupted data in comparison to Yahoo's CSV file on their website. Also some prices have more than 2 decimal places. This is regarding MSFT so I have not tested other symbols.
 
Quote from funnyguy:

It's seems to be a Yahoo server issue . I have tried two independent Yahoo data downloaders both showing same corrupted data in comparison to Yahoo's CSV file on their website. Also some prices have more than 2 decimal places. This is regarding MSFT so I have not tested other symbols.

FYI, it's neither a Yahoo server issue nor a bug in both competing softwares I've used. Both downloaders by default download split and dividend adjusted prices from Yahoo Historical. Then the according main programs by default use adj. close column and automatically adjust High Low Open and Volume fields according to split factor (ADJ.Close/Close) so you have split-adjusted history that is desirable in most cases. That was new to me.

In case of Amiquote if you don't wanna download and import adjusted prices but rather raw (unadjusted) data then you would need to edit

"aqh.format" file (in AmiBroker/Formats subfolder) and change the line:
$FORMAT Date_DMY,Open,High,Low,Close,Volume,AdjClose

to

$FORMAT Date_DMY,Open,High,Low,Close,Volume

After new download and import reload the entire history.

Got the info from support. So everything is in line.
 
Quote from intradaybill:

Your first code does not work well in version 5.40.

Example: On 03/22/2012 Entry on close: 32 - correct

Exit should be on open of 03/23/2012 at 32.10

Instead it exits at 31.98


So you call me a faker all the time although I have provided visual proofs and code proofs. I don't have a problem with other words you mentioned, I'm not a cry boy since I'm not an angel too. But no one who doesn't have any clue at all should call me a faker. If someone calls me a faker then it really gets personal. Then figuratively speaking I transform into a pitbull who won't let go. Sorry budd, but you have made the wrong guy getting annoyed.

I have proven you wrong again. So who is real faker here? The funny thing is you still have not mentioned what you are after. It's absolutely funny that you can not describe a problem in detail. Instead you claim something then someone kindly asks you what it's about. You say "No, you show me". That's ridiculous. What's to show?
 
Hey moron, the discrepancy was identified by looking at the chart of Amibroker and the backtest at the same time. So it does not matter what yahoo downloader does. The Amibroker backtest should use data from the chart you moron. If the data from the chart do not verify the backtest you are ph*ked. So stop trying to confuse people with downloader bs.

The fact of the matter is that I run across this problem long time ago and I did not remember the details. However, now, after your panic, I started working on this and I was able to identify at least one of the problems and that is serious. It is a rather nasty incompatibility between some otherwise irrelevant functions in AFL (not the ones mentioned before). I do not know how this incompatibility may affect the logic of other systems but it is serious. There is one unrelated function to the code I have forgotten at the top from another example I was running because I thought it was irrelevant and it should be. But the presence of that function messes up the results of the code you gave. This is an example how:

attachment.php


The presence of this otherwise irrelevant function causes the discrepancy in the output that I was not able to resolve then some time ago because I just did not spend the time. You can see that positions are held for two days and they are exited at the close of the second day, not at the open of next day. The example is for SPY because it has no splits.

So make me a favor before jumping like a monkey around to pay attention to what someone has to say. I will not reveal the cause of this problem to you unless you sincerely ask for forgiveness. I am telling you, the function that was left in the code on the top was supposed to be irrelevant to the code. This is serious.
 

Attachments

Quote from intradaybill:

Hey moron, the discrepancy was identified by looking at the chart of Amibroker and the backtest at the same time. So it does not matter what yahoo downloader does. The Amibroker backtest should use data from the chart you moron. If the data from the chart do not verify the backtest you are ph*ked. So stop trying to confuse people with downloader bs.

The fact of the matter is that I run across this problem long time ago and I did not remember the details. However, now, after your panic, I started working on this and I was able to identify at least one of the problems and that is serious. It is a rather nasty incompatibility between some otherwise irrelevant functions in AFL (not the ones mentioned before). I do not know how this incompatibility may affect the logic of other systems but it is serious. There is one unrelated function to the code I have forgotten at the top from another example I was running because I thought it was irrelevant and it should be. But the presence of that function messes up the results of the code you gave. This is an example how:



The presence of this otherwise irrelevant function causes the discrepancy in the output that I was not able to resolve then some time ago because I just did not spend the time. You can see that positions are held for two days and they are exited at the close of the second day, not at the open of next day. The example is for SPY because it has no splits.

So make me a favor before jumping like a monkey around to pay attention to what someone has to say. I will not reveal the cause of this problem to you unless you sincerely ask for forgiveness. I am telling you, the function that was left in the code on the top was supposed to be irrelevant to the code. This is serious.

The remark regarding Yahoo data was just a side note and has got nothing to do with the code working right or not! I had to use raw data because you have used raw data of MSFT and I had to verify your one day result (03/22/2012) so I had to re-import raw data because my data of MSFT was split adjusted data! Do you understand that or am still talking to a child?

Again, the code I've provided is working correctly as seen in the video! It correctly outputs the entries and exits in 5.4 and 5.5. And of course AB uses the same data as the one being charted. There is only one data history per symbol! Jesus

I don't care what code you are using. You can tell whatever you want since no one can verify your result without having the same information. Who knows whether the code is wrong or not. On the other side you can't even handle 5 simple lines of code and call them "being complicated". lol So who knows what badly coded stuff you have there. What matters is that I have provided a code according to your description and that code outputs correct results.

If your so called function is such a secret thing then send it to Tomasz. Why haven't you asked support so far? I don't get it. They are much smarter than you will ever be.
 
So here is another video. This time including chart and using symbol SPY. (Apologies for the absolute miserable video quality.)

uGqCNk


I guess next time your majesty wants to see a pony in the video
 
Quote from funnyguy:

If your so called function is such a secret thing then send it to Tomasz. Why haven't you asked support so far? I don't get it. They are much smarter than you will ever be.

Why? Because although I recommend Amibroker to newbies bcause of low cost I don't use it because of potential problems. I don't want to have to verify backtests manually all the time because I have seen function incompatibilities in the software in the past.

Nothing is secret. I will disclose the findings whenever I feel like. I do not jump around like a monkey. You blew a private message you initiated out of proportions of a problem I have encoutered in the past and I really did not pay attention to it. You continuing ad hominen attacks about my abilities won't do you any good. You are a monkey and you will remain one all your life unless you get some human education and related behavior.

My findings about Amibroker will be publicky disclosed in another thread when I like not when you like. I have other things I am doing much more serious than dealing with a backtester with conflicting functions.
 
Back
Top