I have two time series data sets. One with treatment and one without treatment. I want to test if the slope of the two data sets are significantly different.
Which statistical analysis do I use for this?
$\begingroup$
$\endgroup$
2
-
$\begingroup$ Maybe something like the R package Causal Impact: google.github.io/CausalImpact/CausalImpact.html $\endgroup$– william3031Nov 6, 2019 at 4:38
-
$\begingroup$ I’d look at Mann-Kendall analyses. $\endgroup$– EngrStudentJul 10, 2020 at 2:48
Add a comment
|
2 Answers
$\begingroup$
$\endgroup$
If you are willing to go that far, you might fit bivariate state-space models with a common slope and two individual slopes, and compare the fits. Depending on what your data looks like (same units?, same time data points?) you might test the differences for a trend. I think your question is too general to venture a more precise answer.
$\begingroup$
$\endgroup$
By plotting them you can get some idea on how different the data is. Apply SMA on both data and then plot to compare.
