2
$\begingroup$

Can I use "years" as a continuous variable ("years" as calendar years from 1984 to 2014) to see if NDVI (normalized difference vegetation index), of the same area at the same time (summer), has changed positively or negatively over the years, for example with a GLMM or GLM?

$\endgroup$
2
  • $\begingroup$ If you had stock index prices hour-by-hour, you probably wouldn't be asking. In real data, 'continuous' variables must be rounded in some way. Yearly data are more severely rounded than hourly data, and it is a matter of judgment whether yearly data are too 'coarse' to be useful. // However, notice the the point made in the link that it may be convenient to have a 'zero' year and give other years relative to that year. $\endgroup$ – BruceET Sep 3 '20 at 7:07
  • 1
    $\begingroup$ The possibilities are endless. I've seen applications in which distinct years are in essence best treated as discrete or categorical and applications in which fractional years made sense (e.g. 2020.25) for data indexed by say daily dates or monthly dates within a year. With environmental data I've often found fraction of year useful for looking at seasonality. $\endgroup$ – Nick Cox Sep 3 '20 at 8:20
0
$\begingroup$

Yes, there is no GLM assumption that would preclude that

$\endgroup$

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