In R I have used lm()to fit a model. Then I use the confint() function to learn more about the slope. How to I understand the following result? Is the following conclusion true:

With 95% probability the slope is between 1232.44 and 1259.912?

> confint(fit1)
              2.5 %    97.5 %
(Intercept) 14161.38 14649.096
car.sales     1232.44  1259.912
  • 2
    $\begingroup$ You may find it helpful to look at Cross Validated SE where there has been extensive discussion of the interpretation of confidence intervals, eg here (stats.stackexchange.com/questions/26450/…) $\endgroup$ – Adam Bailey Jan 16 '17 at 9:36
  • 3
    $\begingroup$ I'm voting to close this question as off-topic because this is a basic statistics question with no economics content. $\endgroup$ – BKay Jan 16 '17 at 10:26

Unfortunately not, as in frequentist statistics the true slope is not a random variable. This leads to a bit of a mouthful as the conclusion

What you could say is that if your assumptions are correct (such as a linear relationship with random errors) and you repeated the exercise with new sample data and new regressions, $95\%$ of the confidence intervals you would generate for the slope would contain the true value

So you have one interval from a procedure which generates correct intervals $95\%$ of the time

|improve this answer|||||

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