0
$\begingroup$

I am writing a thesis about predicting different series in the FRED-MD database. Before estimating the proposed model, I make the following line plot in R , it presents how the variables in the first group vary over time after making proper transformations. I wonder if this would be a good idea to make such visualisation and how can I interpret from such a line graph.

enter image description here

$\endgroup$

1 Answer 1

0
$\begingroup$

At this zoomed out level, it is impossible to discern the individual lines, and it mostly looks like a mess. I recommend showing a zoomed-in plot in x, where the reader can understand what you're trying to convey. Another trick is showing the different lines at various y-offsets, like 5,10,15,... And explain this in the caption.

$\endgroup$
2
  • $\begingroup$ Thanks, would a correlation plot instead provide better insight as the data contains many variables? $\endgroup$ Jun 20, 2020 at 21:03
  • $\begingroup$ I'm not sure what you mean by a correlation plot, but it really depends on what you want to communicate to the reader. Different types of plots are useful for different purposes. $\endgroup$ Jun 23, 2020 at 1:23

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.