venter {modeest} | R Documentation |
This function computes Venter mode estimator, also called Dalenius, or LMS (Least Median Square) mode estimator.
venter(x, bw = NULL, k, iter = 1, type = 1, tie.action = "mean", tie.limit = 0.05) shorth(x, ...)
x |
numeric. Vector of observations. |
bw |
numeric. The bandwidth to be used. Should belong to (0, 1]. See 'Details'. |
k |
numeric. See 'Details'. |
iter |
numeric. Number of iterations. |
type |
numeric or character. The type of Venter estimate to be computed. See 'Details'. |
tie.action |
character. The action to take if a tie is encountered. |
tie.limit |
numeric. A limit deciding whether or not a warning is given when a tie is encountered. |
... |
Further arguments. |
The modal interval, i.e. the shortest interval among intervals containing k+1
observations, is first computed.
The user should either give the bandwidth bw
or the argument k
,
k
being taken equal to ceiling(bw*ny) - 1
if missing.
If type = 1
, the midpoint of the modal interval is returned.
If type = 2
, the floor((k+1)/2)
th element of the modal interval is returned.
If type = 3
or type = "dalenius"
, the median of the modal interval is returned.
If type = 4
or type = "shorth"
, the mean of the modal interval is returned.
If type = 5
or type = "ekblom"
, Ekblom's L_{-infinity} estimate is returned, see Ekblom (1972).
If type = 6
or type = "hsm"
, the half sample mode (hsm) is computed, see hsm
.
A numeric value is returned, the mode estimate.
The user should preferentially call venter
through
mlv(x, method = "venter", ...)
.
This returns an object of class mlv
.
Paul Poncet paulponcet@yahoo.fr
mlv
for general mode estimation, hsm
for the half sample mode
library(evd) # Unimodal distribution x <- rgev(1000, loc = 23, scale = 1.5, shape = 0) ## True mode gevMode(loc = 23, scale = 1.5, shape = 0) ## Estimate of the mode venter(x, bw = 1/3) M <- mlv(x, method = "venter", bw = 1/3) print(M) plot(M, xlim = c(20, 30))