getRiceParam: Estimate Rice parameters nu and sigma
Description
Estimates the location parameter nu and the scale parameter sigma of the Rice distribution together with the radial mean MR and radial standard deviation RSD based on a set of 2D-coordinates. Includes the parametric confidence interval for sigma.
A list with the estimates for nu, sigma, RSD, and MR including the confidence interval for sigma.
nu
The estimated location parameter nu.
sigma
A vector with the sigma estimate and confidence interval bounds as named elements sigma, sigCIlo, sigCIup.
MR
The MR estimate.
RSD
The RSD estimate.
Arguments
xy
either a numerical (n x 2)-matrix with the (x,y)-coordinates of n points (1 row of coordinates per point), or a data frame with either the variables x, y or point.x, point.y.
level
a numerical value with the coverage for the confidence interval for sigma.
doRob
logical: use robust estimation of center and covariance matrix as basis for estimators?
type
The initial biased nu estimate is the Euclidean norm of the group center. For 'LiZhangDai', the bias-correction from Liu et al., 2009. For 'MOM', the estimated bias is subtracted. If the estimated bias is larger than the initial estimate, the final estimate is then set to 0.
Details
The sigma estimate uses getRayParam. The robust estimate for the center and for the covariance matrix of (x,y,z)-coordinates is from covMcd using the MCD algorithm.
Li, Q., Zhang, J., & Dai, S. (2009). On estimating the non-centrality parameter of a chi-squared distribution. Statistics and Probability Letters 79, 98-114.