Take the 2-minute tour ×
Quantitative Finance Stack Exchange is a question and answer site for finance professionals and academics. It's 100% free, no registration required.

I am working on a paper where I have to model the long run relationship between earnings and dividends. I have downloaded the raw data from shillers website. I have converted the series to log(dividend) and log(earnings), and tested for stationarity in both variables. Both are non stationary before differencing once. The result from the test of cointegration between the variables conclude that they are cointegrated. Studying the raw data (levels without log), both variables have an exponential growth, but after using log there is a linear relationship. I have run a regression with both log and levels but I do not understand which of the models that are correct when describing the long run relationship between the variables.

  1. Log(dividend) = a + b(log(earnings)) + u --> -0.44 + 0.88(log(earnings))

  2. Dividend = a + bEarnings + u --> 0.87 + 0.35Earnings

From the book I am using it is stated that i do not need to work with stationary variables when modelling the long run relationship between two cointegrated variables, and that cointegration allows me to use levels (where the variables separately are non stationary). But when both variables possess an exponential growth, am I supposed to use equation 1 or 2? And Why choose one model over the other one? Also is there a problem with omitted variable bias when modelling the long run relationship between two cointegrating variables?

share|improve this question
    
Many tech stocks would counter the notion that high earners pay lots of dividends. It would seem there's more to predicting dividends than just earnings. –  chrisaycock Apr 11 '13 at 11:30
    
I agree, but the assignment we are given focus on just earnings on dividends, and we are not supposed to include any more variables when modelling the long run relationship. We are also supposed to model an error correction model, where we include lagged values of dividends as well as the lagged laves of the residuals coming from one of the two models described above. Looking at the secong model above, that model states that 1 percent increase in earnings will lead to a 0.87% increase in dividends, and i believe that is too much - and maybe that are caused by omitted variable bias if that is a –  Fred Apr 11 '13 at 12:41
    
problem when modelling the long run relationship between two cointegrating variables. –  Fred Apr 11 '13 at 12:42
    
I think you are misreading your own regressions? –  experquisite Apr 11 '13 at 23:10
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.