how to interpret time series regression

rbcc

Junior Member
Joined
Nov 18, 2009
Messages
126
Hi,

How do interpret regression results with time series data?

If the period covered is 20 years take for example this regression

\(\displaystyle y_t=\beta_0+\beta_1(y)+\epsilon_t\)

if the coefficient on \(\displaystyle \beta_1\) is 500 for example how would I interpret that? Is it the same as a standard regression? A 1 unit increase in \(\displaystyle \beta_1\) is associated with a 500 unit increase in \(\displaystyle y_t\)?


Thank you!
 
Hi,

How do interpret regression results with time series data?

If the period covered is 20 years take for example this regression

\(\displaystyle y_t=\beta_0+\beta_1(y)+\epsilon_t\)

if the coefficient on \(\displaystyle \beta_1\) is 500 for example how would I interpret that? Is it the same as a standard regression? A 1 unit increase in \(\displaystyle \beta_1\) is associated with a 500 unit increase in \(\displaystyle y_t\)?


Thank you!
If I saw this regression formula I would interpret it as y is the independent variably; yt is the dependent variable; \(\displaystyle \beta_0\) and \(\displaystyle \beta_1\) are the linear regression fit variables for the 'explained behavior'; and \(\displaystyle \epsilon_t\) is a measure of the 'unexplained behavior'. Thus a one unit increase in y is associated with a 500 unit increase in \(\displaystyle y_t\)

It really doesn't make any difference what y and yt are. They could be a time dependent behavior and time or number of grains of sand in a swimming suit after a dip in the ocean and size of the swimming suit.
 
Last edited:
Top