Hi TJ

Been thinking about this.

I think I did the original import from a  Metastock database.
The zero(0) closes occur at the beginning of the symbols only ( 99.9% of the 
time ) - so some of the symbols might have 5 - 40 ( it varies a bit ) bars of 
EOD data at the beginning.

All Price fields are set to 0

Only rarely is there a day in the middle of the sequence with a zero close.

In the sequences where there is insufficient data this is perfectly 
understandable, and stdev() and AB seem to give me rational results.

I don't seem to be having any other problems - do you think there might be a 
data corruption issue?

BTW: AB 5.20, XP, 

Regs

RZ







________________________________
From: Tomasz Janeczko <[email protected]>
To: [email protected]
Sent: Sat, 27 March, 2010 2:22:09 AM
Subject: Re: [amibroker] problems with stdev?

  
Hello,

You should not have zero as closing price when you use regular import
funciton
because they don't allow that.

Maybe you simply don't have 200 bars of data for certain symbols.
Or value passed to log function is equal to 1 (no change in price).
Log( 1 ) is equal zero.

Best regards,
Tomasz Janeczko
amibroker.com

On 2010-03-26 15:36, i cs wrote: 
>  
> 
>That
>was quick - thanks herman.
>
>>The problem was a divide by zero!
>
>>Why this got past me was that the zeroes in the closing price seemed to
>occur well before number of bars required for the actual calculation .
>
>>I will look up stdev in the group anyway, sometimes it gets a bit
>frustrating looking through a couple of hundred posts and not findind
>the answer!
>
>>Live and learn!
>
>>Thanks again
>
>>RZ
>
>
>
>
>
>
________________________________
From: >Herman <psy...@magma. ca>
>To: ics4mer
><amibro...@yahoogrou ps.com>
>Sent: Sat, 27 March,
>2010 1:31:17 AM
>Subject: Re:
>[amibroker] problems with stdev?
>
>  
>This was discussed some time back and has a logical explanation.
>Do a
>>search for StDev(
>
>>herman
>
>>> Hi,
>
>>> Getting a 0.0000 during explorations with stdev. 
>>> Not all stocks, just some.
>
>>> However, using the same code, I get successful plots.
>
>>> Can't find anything in the manual to indicate why this is
>>> happening, and there seems to be no odd data in the symbol which
>might cause it to go astray.
>
>>> Here is the code, with some debug statements:
>
>>> VSPR = (C/Ref(C,-1) ); //works OK
>>> VSLog = log(C/Ref(C, -1)); // works OK
>>> VolStd = StDev(log(C/ Ref(C,-1) ),200); // Fails, gives 0.000000
>
>>> Anyone have similar probs?
>
>>> TIA
>
>>> RZ
>
>>> ------------ --------- --------- ------
>
>>> **** IMPORTANT PLEASE READ ****
>>> This group is for the discussion between users only.
>>> This is *NOT* technical support channel.
>
>>> TO GET TECHNICAL SUPPORT send an e-mail directly to 
>>> SUPPORT {at} amibroker.com
>
>>> TO SUBMIT SUGGESTIONS please use FEEDBACK CENTER at
>>> http://www.amibroke
>r.com/feedback/
>>> (submissions sent via other channels won't be considered)
>
>>> For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
>>> http://www.amibroke
>r.com/devlog/
>
>>> Yahoo! Groups Links
>
>
>> 
>
> 
 


      

Reply via email to