I would like to write a flexible programme that excludes observations with particular characteristics. For example, I want to temporary excludes some observations belonging to a series where some variables take too high or too low values (eg for several independent series with 100 realisations I want to exclude the 1st and 99th percentiles of each series).
Unfortunately the easier solution a ‘if/then/else statement’ command cannot be used with series.
Thank you,
Marco
PS I am using e-views 6.
alternative to if-statements with series
Moderators: EViews Gareth, EViews Moderator, EViews Jason, EViews Matt
-
marco.stringa
- Posts: 3
- Joined: Wed Aug 11, 2010 9:58 am
-
startz
- Non-normality and collinearity are NOT problems!
- Posts: 3796
- Joined: Wed Sep 17, 2008 2:25 pm
Re: alternative to if-statements with series
You want smpl if, which specifies a condition for including observations. For example
Code: Select all
smpl if X>-1 and X<17-
marco.stringa
- Posts: 3
- Joined: Wed Aug 11, 2010 9:58 am
Re: alternative to if-statements with series
Thank you very - sorry i thought I could not use the "if" with series in any case.
Who is online
Users browsing this forum: No registered users and 2 guests
