I want to learn the statistics package gretl.

My first attempt to do so is to calculate a linear regression model of a set of data:

$$y_i = \alpha + \beta x_i + u_i$$

First I want to create a crossplot of the data and then calculate the variance of the residuals $s^2_u$ and also those of $\alpha$ and $\beta$.

My questions are:

How to get the crossplot, I do not see any option in the menu which says crossplot? Is this possible in gretl?

When I click Summary statistics I think I get the $s^2$ of the data, however is it also possible to get the the variance of the residuals $s^2_u$ and also those of $\alpha$ and $\beta$?

Is it also possible to calculate in gretl a model without $\alpha$, such as: $$y_i = \beta x_i + u_i$$

I appreciate your answers!


My gretl window:

enter image description here


1 Answer 1


I think this is a stack overflow question and not a Stats question but here goes...

It is possible to get a model without an intercept $\alpha$; simply exclude the const from your X values when you define your model.

To get the variance of the residuals (and a plot of them) try within the model window to go Test > Normality of Residuals.

I don't understand what exactly you mean by crossplot, do you mean fitted against estimated values? (In which case would be Graphs > Fitted, Actual Plots > Actual vs. Fitted.

  • 3
    $\begingroup$ Glad I could help. Next time you ask for something in relation to a software package, try to state the version you use also (Help > About Gretl); it will save you hassle in the long run. $\endgroup$
    – usεr11852
    Apr 2, 2013 at 13:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.