Estimate a density on the cylinder

Estimates a joint or conditional kernel density for circular-linear data on S1 x R. The circular component uses von Mises weights and the linear component uses a Gaussian kernel.

Usage

estimate_cyl_density(
  theta,
  x,
  n_theta = 181,
  n_x = 200,
  kappa = 20,
  h = NULL,
  x_grid = NULL,
  conditional = TRUE
)

Arguments

theta

Angles in radians.

x

Real-valued response.

n_theta

Number of angular grid points.

n_x

Number of linear grid points.

kappa

Circular concentration for the conditioning kernel.

h

Linear bandwidth. If NULL, a rule-of-thumb bandwidth is used.

x_grid

Optional grid for the linear response.

conditional

If TRUE, rows are normalized to estimate f(x | theta).

Value

A list with grids, a density matrix, a long data frame and smoothing metadata.

See also

Other cylindrical dependence helpers: circular_topography_data(), phase_loom_data(), stat_orbit_data()

Examples

dat <- simulate_cylindrical(n = 80, seed = 1)
est <- estimate_cyl_density(dat$theta, dat$x, n_theta = 24, n_x = 30)
str(est$data)
#> 'data.frame':	720 obs. of  3 variables:
#>  $ theta  : num  0.131 0.393 0.654 0.916 1.178 ...
#>  $ x      : num  1.41 1.41 1.41 1.41 1.41 ...
#>  $ density: num  1.11e-10 2.88e-10 8.77e-10 1.79e-09 1.49e-09 ...
#>  - attr(*, "out.attrs")=List of 2
#>   ..$ dim     : Named int [1:2] 24 30
#>   .. ..- attr(*, "names")= chr [1:2] "theta" "x"
#>   ..$ dimnames:List of 2
#>   .. ..$ theta: chr [1:24] "theta=0.1308997" "theta=0.3926991" "theta=0.6544985" "theta=0.9162979" ...
#>   .. ..$ x    : chr [1:30] "x=1.413491" "x=1.654167" "x=1.894842" "x=2.135518" ...