The problem starts with the AB WMA assigning values to the empty spaces in
WMA values at the beginning of the data series. It keeps doing that until
there are approximately WMA period * (Expand Period + 1) bars available
which is apparently considered the 1st bar with good or usable data. In
addition I really don't understand why so many bars are required: One would
think that to plot the first expansion bar only (WMA Period + Expansion
Period) bars are required. Also the expandperiod itself seems to be off:
When selecting expandperiod =n it expands n+1. This is easy to check with
the new XBar indicator of the cross Hair. It seems a typical case of 5
trees and 5 spaces. Below an AFL which demonstrates it in grinding detail.
Before I switched over for compatibility, when it became available, to TJ's
Timeset/restore/expand stuff I used my own time compressed formulas. Thise
had the additional advantage that I could select a rolling week, month what
provided me with half an interval earlier data. However, I have no idea on
all my hard drives where they are. All I know I have them, the question
is:where?
/*
1. Click first on WMAFormVerON|Off parameter to see the difference between
the common WMA and the AB-WMA
2. Thereafter turn the parameter off and play with the WMAON_Off parameter
with various periods
3. For comparison an EMA is included.
4. Minimum # of bars required = WMA period * (Expand Period + 1) bars
(approximately).
*/
Title = EncodeColor(4)+ _DEFAULT_NAME()+"; "+EncodeColor(1) +
StrFormat("{{NAME}} - {{INTERVAL}}; {{DATE}}; O=%g, H=%g, L=%g, C=%g
(%.1f%%)
{{VALUES}}", O, H, L, C, SelectedValue( ROC( C, 1 ) ) );
WMAFormulaVerON_Off=ParamToggle("WMAFormVerON|Off","WMAVerON|Off",1);
WMAON_Off=ParamToggle("WMA ON|Off","WMAON|Off",1);
MAPeriod=Param("MAPeriod",9,1,40,1);
ExpandPeriod=Param("ExpandPeriod",5,1,100,1);
///////////////////////////WMA Formula
Verification////////////////////////////////////////
if (WMAFormulaVerON_Off==0)
{
C1=Ref(C,-1);C2=Ref(C,-2);C3=Ref(C,-3);C4=Ref(C,-4);
WMASelf5 = (5*C+ 4*C1+ 3*C2+ 2*C3+ 1*C4)/15;
Plot(WMASelf5,"WMA - Joris",colorGreen,5);
Plot(WMA(C,5),"WMA - AB",colorRed,5);
}
///////////////////////////TimeFrameSet/Restore/Expand()
Verification////////////////////////////////////////
if (WMAFormulaVerON_Off==1)
{
TimeFrameSet( inDaily ); // switch now to Daily
maDEMA = EMA( C, MAPeriod); TimeFrameRestore(); // restore time frame to
original
TimeFrameSet( inDaily*ExpandPeriod); // switch now to Daily
maD5EMA = EMA( C, MAPeriod);
TimeFrameRestore(); // restore time frame to original
Plot( Close, "Price", colorBlack, styleCandle );
// plot expanded average
Plot( TimeFrameExpand( maDEMA , inDaily ),"\n"+ WriteVal(MAPeriod,1.0)+"bar
EMA from Daily bars", colorRed );
Plot( TimeFrameExpand( maD5EMA , inDaily*ExpandPeriod),
WriteVal(MAPeriod,1.0)+" bar EMA from Daily*"+WriteVal(ExpandPeriod,1.0)+"
bars", colorGreen,expandFirst );
if(WMAON_Off==0)
{
TimeFrameSet( inDaily ); // switch now to Daily
maD = WMA( C, MAPeriod);
TimeFrameRestore(); // restore time frame to original
TimeFrameSet( inDaily*ExpandPeriod); // switch now to Daily
maD5 = WMA( C, MAPeriod);
TimeFrameRestore(); // restore time frame to original
MinBarsReq = MAPeriod * (ExpandPeriod + 1);
Plot( Close, "Price", colorBlack, styleCandle );
// plot expanded average
Plot( TimeFrameExpand( maD, inDaily ),"\n"+ WriteVal(MAPeriod,1.0)+" bar WMA
from Daily bars", colorBlue );
Plot( TimeFrameExpand( maD5, inDaily*ExpandPeriod),
WriteVal(MAPeriod,1.0)+" bar WMA from Daily*"+WriteVal(ExpandPeriod,1.0)+"
bars", colorViolet );
Plot( TimeFrameExpand( maD5, inDaily*ExpandPeriod), "Minimum bars required =
"+WriteVal( MinBarsReq,1.0), colorViolet );
}
}
From: [email protected] [mailto:[email protected]] On Behalf
Of johncw_97
Sent: Wednesday, September 16, 2009 5:46 PM
To: [email protected]
Subject: [amibroker] Re: WMA problem
--- In [email protected] <mailto:amibroker%40yahoogroups.com> ,
"rkp1" <ralphpatriq...@...> wrote:
>
> Hi folks. I'm having a problem with the WMA function in TimeFrames. When I
use the following code in the default Price chart:
>
> TimeFrameSet(inDaily * 5);
> mw = WMA(C, 39);
> TimeFrameRestore();
> mwx = TimeFrameExpand(mw, inDaily * 5);
> Plot(mwx, "WMA", colorRed, styleThick);
>
> I get 4 {EMPTY} bars followed by almost 200 bars starting at
-9,500,000,256.00 to -12,820,173.00 before the WMA settles down. Could
others verify this, and is there a way to eliminate it?
>
> Thanks,
>
> Ralph
>
About 6 years ago, I had the same experience. My workaround
is an AFL function wrapper for WMA() so that when used with
TF functions, produces what I expect. I am not on my own
computer now - email if you want the wrapper and I will send it
to you.
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.409 / Virus Database: 270.13.100/2375 - Release Date: 09/16/09
05:51:00
