init_ivols
find an initial estimate of the intrinsic volumes via
moment-fitting.
init_ivols(d, init_mode = 0, delta = d/2, var = d/4)
d | the dimension of the bivariate chi-bar squared distribution. |
---|---|
init_mode | specifies the way through which the initial estimate is found:
|
delta | an estimate of the statistical dimension of the cone. |
var | an estimate of the variane of the intrinsic volumes. |
The output of init_ivols
is a (d+1)
-column vector.
rbichibarsq
, circ_rbichibarsq
,
rbichibarsq_polyh
, loglike_ivols
Package: conivol
D <- c(5,5) d <- sum(D) alpha <- c(pi/3,pi/4) v_exact <- circ_ivols(D, alpha, product=TRUE) m_samp <- rbichibarsq(10^6,v_exact) est <- estim_statdim_var(d, m_samp) list( v_exact = v_exact , v_init0 = init_ivols( d ) , v_init1 = init_ivols( d, 1, delta=est$delta, var=est$var ) , v_init2 = init_ivols( d, 2, delta=est$delta ) , v_init3 = init_ivols( d, 3, delta=est$delta, var=est$var ) , v_init4 = init_ivols( d, 4, delta=est$delta, var=est$var ) )#> $v_exact #> [1] 0.0007466705 0.0059021152 0.0238862411 0.0650168700 0.1304771451 #> [6] 0.1959118384 0.2206475000 0.1866928441 0.1151708399 0.0464763322 #> [11] 0.0090716034 #> #> $v_init0 #> [1] 0.09090909 0.09090909 0.09090909 0.09090909 0.09090909 0.09090909 #> [7] 0.09090909 0.09090909 0.09090909 0.09090909 0.09090909 #> #> $v_init1 #> [1] 0.001027626 0.005615565 0.022273421 0.064145431 0.134169050 0.203861994 #> [7] 0.225045217 0.180495106 0.105170839 0.044513592 0.013682161 #> #> $v_init2 #> [1] 0.003195552 0.015418752 0.042267282 0.085501810 0.136724769 0.177800371 #> [7] 0.189545420 0.164326460 0.112616650 0.056952606 0.015650327 #> #> $v_init3 #> [1] 0.0003151408 0.0021400851 0.0084990689 0.0249073521 0.0577011768 #> [6] 0.1087064603 0.1678885992 0.2108631213 0.2093538516 0.1533829812 #> [11] 0.0562421628 #> #> $v_init4 #> [1] 0.001066540 0.006105089 0.020143723 0.049045970 0.094399095 0.147756265 #> [7] 0.189591624 0.197836291 0.163190050 0.099333854 0.031531499 #>