The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
density_asymmetry(): the kernel density estimate is now
wrapped around the circle (wrap = TRUE), so mass near +/-
circ_space/2 is no longer truncated. This is a no-op to
~1e-13 for orientation or colour errors, but it matters for data with a
180-degrees-off reversal mode, such as motion direction.density_asymmetry(): the +/- circ_space/2
point is now excluded from the positive and negative sums
(exclude_antipode = TRUE), like zero. It is the same angle
reached from two sides, and with an odd n it was counted
twice.density_asymmetry(): a bandwidth narrower than the grid
spacing now rescales yvar and the bandwidth together
(rescale_narrow = TRUE) so the kernel spans at least one
cell. Without it, the rectangle sum aliases: on a narrow error
distribution with bw.SJ()-selected bandwidth the returned
asymmetry was off by 7.5% of its value at the default
n = 181, now 0.4%.density_asymmetry(): the two signs are now paired on a
rounded abs(x), since seq() is not bitwise
symmetric for every n (on n = 1801, 604 of 900
pairs failed to match and dropped out of the sums entirely).density_asymmetry_discrete(): the same four fixes, with
the same argument names and defaults. Wrapping there is done by
evaluating stats::density() over the neighbouring periods
and folding the images back, since augmenting the sample with shifted
copies would run into density()’s binning range.wrap = FALSE, exclude_antipode = FALSE,
rescale_narrow = FALSE for pre-1.4.0 results.circ_mean(), circ_sd(), and
angle_diff(), generalized versions of the
_rad/_180/_360 family that accept
an arbitrary period argument (e.g., 180, 360, or anything
else).period argument to
weighted_circ_mean() to compute the weighted circular mean
directly on data in an arbitrary circular space, without manually
converting to radians first.remove_cardinal_biases(): replaced gamlss
dependency with a native P-spline + REML/EM implementation. Results are
numerically equivalent; runtime is approximately 2× faster.gamlss from package imports.density_asymmetry_discrete().smoothed_circ_sd().density_asymmetry() speed and options.smoothed_circ_sd().density_asymmetry() by using matrix
multiplication."average" as a bandwidth option in
density_asymmetry().ggplot2, stats) and
documentation.density_asymmetry_discrete().return_full_density option to
density_asymmetry().density_asymmetry() to analyze the asymmetry in
response probabilities.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.