Skip to main content

Table 2 Different predictor combinations compared during AIC-based iSSM and ST-PPM selection

From: Analysis of local habitat selection and large-scale attraction/avoidance based on animal tracking data: is there a single best method?

iSSM predictors ST-PPM predictors  
hab hab  
hab+log(dx) hab+s(log(dx))  
hab+cos(ta) hab+s(cos(ta))  
hab+log(dx)+cos(ta) hab+s(log(dx))+s(cos(ta))  
hab+log(dx)+cos(ta) + log(dx):cos(ta) hab+te(log(dx),cos(ta))  
  1. s(.) defines a cubic regression spline, te(.) a tensor product spline [53], ta is the turning angle, and dx the spatial distance to the foregoing tracking point