I am trying to analyse the results that I get from regressing a variable (in this case investment) on a constant and an independent variable (GDP). When I run my first regression, I use the least-squares command on lagged variables: ls log(Investment) c log(GDP) and I see that there is a considerable serial correlation between these two variable when checking the scatterplot.
However, when I run a second regression taking differentials, using this command: ls dlog(Investment) c dlog(GDP) and I plot the scatterplot, I see that it everything is more spread out and doesn't seem like it's highly correlated. From an analytic point of view, should I really consider the second scatterplot for analysis?
For econometric discussions not necessarily related to EViews.
1 post • Page 1 of 1
Who is online
Users browsing this forum: No registered users and 29 guests