10 Nov
2011
10 Nov
'11
1:51 p.m.
Hi Scott,
The lags and leads arguments only include one lag or lead. Any more lags/leads would need
to be constructed in the data before imputation.
Cheers,
matt.
~~~~~~~~~~~
Matthew Blackwell
PhD Candidate
Institute for Quantitative Social Science
Department of Government
Harvard University
url: http://www.mattblackwell.org
On Monday, November 7, 2011 at 3:38 AM, Scott Otterson wrote:
Hello,
When Amelia is given a lags or lead argument, does it include in its model only one lag
or lead, advanced or delayed by one sample? Or does it instead consider multiple lags or
leads and pick the best subset with some algorithm?
Thanks,
Scott
Dr. Scott Otterson
Abteilung Energiemeteorologie und Netzintegration
Fraunhofer IWES
Tel: +49 (0)561 7294-252
E-Mail: scott.otterson(a)fraunhofer.iwes.de (mailto:scott.otterson@fraunhofer.iwes.de)