You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello!,
I have been using BEAST to detect change points in Sensor data,
specifically indices such as NDVI, NDWI, NDBI, and surface temperature.
The data is seasonal from 1987-2022. Aprox 130 obs.
Contrary to what I expected,
the algorithm do not detect more than one breakpoint. For more of the indices, it detects 0 or 1 changepoints
This do not occurred with the temperature index, where at least 2 or 3 points are detected.
Does this has something to do with the input parameters?
We even convert dates to fractional dates.
NDVI, NDWI, NDBI are normalized indices, [-1;1] does this affect the results? - Is it possible to obtain the direction of the slope? Due to BEAST ouyput non-linear trends?
Hello!,
I have been using BEAST to detect change points in Sensor data,
specifically indices such as NDVI, NDWI, NDBI, and surface temperature.
The data is seasonal from 1987-2022. Aprox 130 obs.
Contrary to what I expected,
the algorithm do not detect more than one breakpoint. For more of the indices, it detects 0 or 1 changepoints
This do not occurred with the temperature index, where at least 2 or 3 points are detected.
Does this has something to do with the input parameters?
We even convert dates to fractional dates.
- Is it possible to obtain the direction of the slope? Due to BEAST ouyput non-linear trends?
Thank you in advance!
The text was updated successfully, but these errors were encountered: