Hi all,
I am having trouble understanding how to interpret the standard error of
the regression (aka residual standard error, in R lingo). In section
notes, it says that we can analyze the meaning of our standard error of
the regression in comparison with the range, standard deviation, min and
max of the dependent variable. But how are we comparing these? Are we
looking for values below, similar to or above these values? I feel like
there is a conceptual link that I am having trouble making here.......
Also, I am not sure in problem three how to "examine and interpret" (Fox,
96) the intercept. I understand the information the slop is giving us but
what does the intercept tell us about the relationship between x and y
suggested by the regression line?
Thanks,
Becky
Hi Everyone,
New versions of the lecture slides are now available on the course
website. The slides for today's lecture have been revised a bit.
Best,
Kevin
------------------------------------------------------
Kevin Quinn
Assistant Professor
Department of Government and
Center for Basic Research in the Social Sciences
34 Kirkland Street
Harvard University
Cambridge, MA 02138
> One more general question for all problems where we have to
> comment on how well the line captures the relationship between the
> variables: does this mean we have to comment on the underliying
> assumptions about variance, independence, etc, or do you expect us to
> generally say whether the point cloud is linearly shaped?
I expect your descriptions to vary by question. For example, in 1c, I
expect the more general assessment you describe. In 7a, on the other
hand, we ask you to analyze specific statistics that assess model fit.
For Problem 3, Fox 5.8, at least provide a general assessment of fit, and
note any strong deviations from the assumptions that you observe.
Assessing all the assumptions, recording R^2 and the SE of the regression,
AND providing a general assessment of fit for each of the three
regressions would be the best possible answer, but at least give a general
assessment, and comment on any strong deviations that inform your general
assessment.
Hope this is helpful,
Ryan