BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
sasgiaz
Quartz | Level 8

Hi. I'm analyzing trend effect on my data which has hundred million rupiah as its unit. 

I got the result of the parameter estimation for the interceipt valued more than hundred million rupiah and I think it's weird and doesn't make sense.. Because I think the result will be around hundred million rupiah like the original data valued. Can anyone tell me what is wrong with my analysis? I use the syntax like this: 

 

proc reg data = arimaxvlp;
model Transaction = Time/influence;
run;

 

Thank u so much for ur help.

1 ACCEPTED SOLUTION

Accepted Solutions
PGStats
Opal | Level 21

It could be that time=0 (where the intercept is estimated) is very far away from the time values in your data. You could center your time somewhere within your time data range to get a more meaningful intercept estimate.

PG

View solution in original post

2 REPLIES 2
PGStats
Opal | Level 21

It could be that time=0 (where the intercept is estimated) is very far away from the time values in your data. You could center your time somewhere within your time data range to get a more meaningful intercept estimate.

PG
sasgiaz
Quartz | Level 8

I understand. Thank you so much!