SPSSX Discussion

using prior years to compute new variable

Classic

List

Threaded

2 messages Options

wall

using prior years to compute new variable

Dear all,

it sounds easy, but I cannot come up with a solution.

I have the number of inhabitants for all cities for all years except for 2011:

2007 2008 2009 2010 2011
12187,0 12131,0 12114,0 12032,0 .
10601,0 10550,0 10462,0 10447,0 .
9514,0 9497,0 9270,0 9261,0 .
7246,0 7276,0 7447,0 7427,0 .
27653,0 27314,0 26744,0 26635,0 .
181613,0 182484,0 184227,0 187298,0 .

and now I want to compute the values for 2011 by extending the regression line for all prior years up to 2011.

For another variable I have another problem. There is a year missing in the middle.

2007 2008 2009 2010
12187,0 12131,0 . 12032,0
10601,0 10550,0 . 10447,0
9514,0 9497,0 . 9261,0
7246,0 7276,0 . 7427,0
27653,0 27314,0 . 26635,0
181613,0 182484,0 . 187298,0

I there a way to interpolate by making use of all other years. At the moment I have used the average of 2008 and 2010 to make the variable for 2009. But I hope there is a better solution which makes use of all information i have.

Thanks for any help!

David Marso

Re: using prior years to compute new variable

Administrator

VARSTOCASES
SPLIT FILE
REGRESSION (save PRED) OR RMV (see options -TREND?-) or CURVEFIT or ....???
CASESTOVARS.

wall wrote

Dear all,

it sounds easy, but I cannot come up with a solution.

I have the number of inhabitants for all cities for all years except for 2011:

2007 2008 2009 2010 2011
12187,0 12131,0 12114,0 12032,0 .
10601,0 10550,0 10462,0 10447,0 .
9514,0 9497,0 9270,0 9261,0 .
7246,0 7276,0 7447,0 7427,0 .
27653,0 27314,0 26744,0 26635,0 .
181613,0 182484,0 184227,0 187298,0 .

and now I want to compute the values for 2011 by extending the regression line for all prior years up to 2011.

For another variable I have another problem. There is a year missing in the middle.

2007 2008 2009 2010
12187,0 12131,0 . 12032,0
10601,0 10550,0 . 10447,0
9514,0 9497,0 . 9261,0
7246,0 7276,0 . 7427,0
27653,0 27314,0 . 26635,0
181613,0 182484,0 . 187298,0

I there a way to interpolate by making use of all other years. At the moment I have used the average of 2008 and 2010 to make the variable for 2009. But I hope there is a better solution which makes use of all information i have.

Thanks for any help!

Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"