09 may 3.dos. Timing useful and you can experience towns
I checked five housing density groups aggregated within a 3 hundred yards scale; 0 properties each ha (wildland), 0 to a single.47 (outlying and you may exurban), 1.47 to ten (suburban), and greater than 10 home for each and every ha (urban). Hr of the GPS neckband metropolises try discretized into the four bins: night (–), morning (–), big date (–), and nights (–) time periods. For every single private and you can period of time, i summed what number of GPS collar location findings from inside the each construction occurrence classification. To standardize the brand new proportional utilization of the construction densities round the big date (as schedules are very different in the course), i next split up the new ensuing worthy of by the final number regarding locations (for the private) one decrease towards whenever bin class.
Presenting the latter, i made use of the inverse logit of your estimated cougar incident regression equations, excluding brand new intercept, to help you aesthetically establish new cousin likelihood of an excellent cougar incident and you can constrain the costs anywhere between no plus one, right after which multiplied each skin of the a surface representing cousin cougar habitat choice
To model the possibility of cougar incidents, we employed a good have fun with?supply build complement having fun with an effective logistic regression (Manly, McDonald, Thomas, McDonald, & Erickson, 2007).