716
or any other type of treatment. It has also been used in hydrology to model stream flow and precłpitation, and in economics to model the distribution of wealth or income.
Kamps [24] introduced the concept of generalized order statistics (gos) as follows: Let Xi,X2... be a seąuence of independent and identically distributed (iid) random vari-ables (rv) with absolutely continuous cumulative distribution function (cdf) F(x) and pdf, f(x), x G (a, /3). Let n G N, n > 2, k > 0, m G 3?, be the parameters such that
7r = k + (n — r)(m + 1)> 0, for all r G {1,2,..., n — 1}, where Mr = YlijZr m3- Then X(l,n,m, k),... ,X(n,n,m, k), r = 1,2,... n are called gos if their joint pdf is given by
on the cone F-1(0) < xi < x% < ... < xn < F-1( 1).
The model of gos contains as special cases, order statistics, record values, seąuential order statistics.
Choosing the parameters appropriately (Cramer, [181), we Set the variant of the gos given in Table 1.
Table 1: Yariants of the generalized order statistics
■4Ć II £ |
Ir |
mr | |
i) Seąuential order statistics |
(n — r + l)or |
I + 1 | |
ii) Ordinary order statistics |
1 |
n — r + 1 |
0 |
ii) Record values |
1 |
1 |
-1 |
iv) Progressively type II censored order statistics |
Rn + 1 |
n-r + 1 + Yjj=r |
Rr |
v) Pfeifer’s record values |
0n |
Pr |
0r ~ Pr+1 ~ 1 |
For simplicity we shall assume mi — mi — ... — mn-1 = m.
The pdf of the i—th gos, X(r,n,m,k), 1 < r < n, is
and the joint pdf of X(r, n, m, k) and X(s,n,m,k), 1 < r < s < n, is
fx(= (r _ Ł)^’ r _ WCWĆ ‘(^l1))
where
F(x) = 1 -F(x), Cr-1 = i li = fc+(n—i)(m+l),
t=i
hm(x) = |_|n(+1_x)j m= — i