proj4, make north up

summary.ctmm CIs on boundary estimates with negative curvature

akde with user-specified grid (and projection)
can we rotate coordinate system to diagonalize sigma?
Eli's kde code: armadillo, eigen, fastest

vignette: gazelle dt, variogram.fit, AKDE errors and interpretation

periodogram:  fast, model, averaging, zoom

north vector in telemetry object

contour labels in plot ctmm

Email ks author

Figure out optimal grid size from bandwidth

Repeated times in variograms -- account for as error -- fix cap

dplyr, rolling for loops

mean.akde

fix mean parameter for boundary issues

population fit of list
population akde of lists

krige -> predict

reverse order of tau so that we have sigma, tau_r, tau_v, tau_a, ... the order of increasing continuity