BDSKY sampling proportion to 0 before the first sample

[Daiana]

Dear All,

For my analyses using BDSKY I want to set the sampling proportion to 0 before the first sample.
To this end I typically edited the XML file by hand adding, among other things, two big modifications:

1) <!-- parameters specifying Zero-Sampling-Before-First-Sample -->

<parameter dimension="2" id="samplingProportion.t:NAME" lower="0.0" name="samplingProportion" upper="1.0" value="0. 0.005"/>
<parameter name="samplingRateChangeTimes" value="0. 11.5"/>
<!-- latest sample time (2014) - earliest sample time (2003) = 11 , the value informed need to be bigger that the calculated -->
<reverseTimeArrays value="false false true false" spec="beast.core.parameter.BooleanParameter"/>
 
     
2)<!-- Example of a ExcludablePrior on the samplingProportion-->
<distribution spec="beast.math.distributions.ExcludablePrior" xInclude="false true" id="samplingProportionPrior" name="distribution" x="@samplingProportion.t:NAME">
<LogNormal name="distr" id="LogNormalDistributionModel.2">
<parameter name="M" id="RealParameter.24" estimate="false">-1.9</parameter>
<parameter name="S" id="RealParameter.25" upper="5.0" lower="0.0" estimate="false">0.05</parameter>
</LogNormal>
</distribution>

However, there is something that is causing an error and I can not identify it.
I attached an image of the output message.

Could anyone help me?

Is there any available template to set this up?

Best regards,

Daiana

[Remco Bouckaert]

Hi Daiana,

The error message indicates that the ExcludabelPrior cannot be accessed. This class is part of the BEASTLabs package. If you install that package (http://www.beast2.org/managing-packages/ for instructions on how to manage packages) the error should not appear any more.

Cheers,

Remco

[Daiana]

Hi Remco,

Thanks a million for your prompt reply!

That was exactly what I was missing. 

Now everything works as expected.

Best regards,

Daiana