Besides showing this in a graph, I was wondering if there is a statistical test i can run to show how the trend changes around that time.
Thinking of using regression with a knot around the timepoint - but I was wondering if there is something better to use.
Thank you in advance good people
Sample data below
Code:
* Example generated by -dataex-. For more info, type help dataex clear input int date float study 21454 1 21525 1 21335 1 21424 1 21222 1 21405 1 21265 1 21283 1 21515 1 21500 1 21399 1 21328 1 21500 1 21458 1 21234 1 21535 1 21682 1 21899 1 21911 1 21682 1 21885 1 21710 1 21871 1 21895 1 21879 1 21774 1 21860 1 21801 1 21895 1 21818 1 21868 1 22174 1 22055 1 22050 1 22117 1 22141 1 22273 1 22070 1 22105 1 22105 1 22259 1 22196 1 22236 1 22001 1 22075 1 21976 1 21935 1 22271 1 22091 1 22250 1 22127 1 22251 1 21985 1 21942 1 21966 1 22494 1 22501 1 22578 1 22512 1 22445 1 22568 1 22305 1 22561 1 22371 1 22421 1 22589 1 22399 1 22460 1 22476 1 22314 1 22592 1 22404 1 22434 1 22582 1 22561 1 22390 1 22307 1 22462 1 22637 1 22419 1 22452 1 22453 1 22587 1 22368 1 22385 1 22579 1 22606 1 22419 1 22307 1 22351 1 22413 1 22295 1 22481 1 22319 1 22490 1 22725 1 22726 1 22768 1 22652 1 22880 1 end format %td date
0 Response to Help with time-trend analysis
Post a Comment