1
$\begingroup$

I'm trying to explain the evolution of productivity(total factor productivity) by its determinants in my country with 46 observations for each variable I explain:

I have the rate of growth of productivty a time series and I'm trying to explain it with the following independent variables:

  • openness to trade
  • political stability
  • inflation rate
  • human capital
  • index
  • credits to private sector
  • foreign investements

These variables are ratios to gdp except the dependant variable which is productivity,and other variables(human capital, inflation rate, political stability).

I found no correlation between the independent variables and the productivity (dependent) variable.

What could be the appropriate model for such a case ?

enter image description here

And that's the correlation matrix and the p-value matrix of correlation.

enter image description here

The dependent variable is ptfg the second column of the data set.

$\endgroup$
  • $\begingroup$ I want just to precise that there is low correlation between the response variable and the dependant variable. and a linear regression doesn't fit well $\endgroup$ – user162339 May 23 '17 at 13:11
  • $\begingroup$ It's impossible for us to give effective advice on which type of model to use with only the information you've provided beyond suggesting a generally time-series appropriate approach like ARIMAX or similar. Some ways to improve our ability to help you with the question would be 1. to post the dataset you're using (or links to it) 2. posting the full results of the correlation calculation you've done including reproducible code etc. The more information we have the better the advice will be. $\endgroup$ – Thomas Cleberg May 23 '17 at 13:41
  • $\begingroup$ I've edited my question @ThomasCleberg $\endgroup$ – user162339 May 23 '17 at 14:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.