I'm attempting to calculate the coefficients of the ARx vegetation stability model as described in Keersmaecker et al's 2016 paper, which I've attached.
The terms in this model are: NDVI at time step t, SPEI (a drought index) at time step t, the temperature anomaly at time step t, NDVI at the previous time step, and the error at time step t. I have all the data I need; I've already calculated NDVI, SPEI, and temperature for my 17-year time series. The coefficients from this equation represent the stability metrics.
I'm attempting to calculate the coefficients in Matlab using the ARx function, but I haven't been able to figure out how to use this function. This model seems relatively simple, such that I could use a multiple linear regression and just manually input the NDVI value from the previous time step.
Even though the literature describes this as an ARx model, would it be inaccurate for me to calculate this as a multiple linear regression? Thanks for any feedback.
The ARX model is a least squares method, which solves linear regression equations. So yes, you can calculate the coefficients as a multiple linear regression :)
- Badges
- Science method