29 Oct
2003
29 Oct
'03
9:16 p.m.
So for 3.c)...does anyone know an easy way of generating our own lags?
(producing a new lag variable)
Zoe
29 Oct
29 Oct
9:52 p.m.
hi zoe:
i suspect there are several ways to do this. However, our group used a for
loop to generate a new variable where each index in the new variable is i+1
of the old variable (you can add more to the index number to make the lag
longer). this is a pretty easy way to do it if you have mastered the for
loop. you lose a small number of values when you do this (but of course,
this is inevitable).
if this is not clear, let me know and i will clarify further.
jason
----- Original Message -----
From: "Zoë VanderWolk" <vanderw(a)fas.harvard.edu>
To: <gov1000-list(a)fas.harvard.edu>
Sent: Wednesday, October 29, 2003 9:16 PM
Subject: [gov1000-list] index numbers
So for 3.c)...does anyone know an easy way of
generating our own lags?
(producing a new lag variable)
Zoe
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
30 Oct
30 Oct
12:03 a.m.
Hello:
You can't really generate a lag variable by itself. Think of it as
generating two new variables, i.e. unrate2, unrate2.lag.
unrate2 would just be data.approval$unrate with the first element removed
(i.e. unrate2 <- data.approval$unrate[2:(length(data.approval$unrate))] )
and unrate2.lag would be data.approval$unrate with the last element
removed ... or vice versa ... maybe I'm getting mixed up, but you get the
idea I hope.
--Andrew
On Wed, 29 Oct 2003, Jason Lakin wrote:
hi zoe:
i suspect there are several ways to do this. However, our group used a for
loop to generate a new variable where each index in the new variable is i+1
of the old variable (you can add more to the index number to make the lag
longer). this is a pretty easy way to do it if you have mastered the for
loop. you lose a small number of values when you do this (but of course,
this is inevitable).
if this is not clear, let me know and i will clarify further.
jason
----- Original Message -----
From: "Zoë VanderWolk" <vanderw(a)fas.harvard.edu>
To: <gov1000-list(a)fas.harvard.edu>
Sent: Wednesday, October 29, 2003 9:16 PM
Subject: [gov1000-list] index numbers
So for 3.c)...does anyone know an easy way of
generating our own lags?
(producing a new lag variable)
Zoe
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
2 Nov
2 Nov
12:43 p.m.
New subject: Question 4b
Hi everyone: My question concerns 4(b) on Homework 5. The information I'm using
to try to answer the question is:
P(support|right)=40%
P(don't support|right)=60%
P(right)=80%
P(wrong)20%
Has anyone had any luck answering this question with this information? Or, are
we to assume that we can use our answer from 4(a) which employs a different
number for the overall support for the war? (is this what is meant by assuming
this observe conditional probability holds?)
any help appreciated.
thanks,
margie
10:03 p.m.
New subject: Question 8
We're assuming that voters will vote either Democratic or Republican
right? Do we have to account for some voters abstaining / voting for
third-party?
-Andrew
10:53 p.m.
New subject: Question 8
We're assuming that voters will vote either
Democratic or Republican
right?
You don't have to make that exact assumption, but this is an okay
assumption to make. If you want to make it, clearly state it and move
on. One needs to make some assumption like it, but there are a number
to choose from (some of which are more general). The logic gets more
complicated for more general assumptions....
JS.
Andrew Calvin Fong writes:
We're assuming that voters will vote either
Democratic or Republican
right? Do we have to account for some voters abstaining / voting for
third-party?
-Andrew
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
7497
days inactive
7501
days old
gov1000-list@lists.fas.harvard.edu
5 comments
5 participants
participants (5)
-
Andrew Calvin Fong -
Jasjeet Singh Sekhon -
Jason Lakin -
jenkins＠fas.harvard.edu -
Zoë VanderWolk