Jeffreys interval. Beta distributions are the
standard conjugate priors for binomial distributions
and it is quite common to use beta priors for inference
on p (see Berger, 1985).
Suppose X ∼ Binn p and suppose p has a prior
distribution Beta a1 a2; then the posterior distribution
of p is BetaX + a1 n − X + a2. Thus a
1001 − α% equal-tailedBayesian interval is given
by
Bα/2 X + a1 n − X + a2
B1 − α/2 X + a1 n − X + a2
where Bα m1 m2 denotes the α quantile of a
Beta m1 m2 distribution.
The well-known Jeffreys prior and the uniform
prior are each a beta distribution. The noninformative
Jeffreys prior is of particular interest to us.
Historically, Bayes procedures under noninformative
priors have a track record of good frequentist
properties; see Wasserman (1991). In this problem
the Jeffreys prior is Beta1/2 1/2 which has the
density function
fp = π−1p−1/21 − p
−1/2
The 1001 − α% equal-tailed Jeffreys prior interval
is defined as
(6) CIJ = LJx UJx
where LJ0 = 0 UJn = 1 and otherwise
(7) LJx = Bα/2 X + 1/2 n − X + 1/2
(8) UJx = B1 − α/2 X + 1/2 n − X + 1/2
The interval is formed by taking the central 1 − α
posterior probability interval. This leaves α/2 posterior
probability in each omitted tail. The exception
is for x = 0n where the lower (upper) limits are
modified to avoid the undesirable result that the
coverage probability Cp n → 0 as p → 0 or 1.
The actual endpoints of the interval need to be
numerically computed. This is very easy to do using
softwares such as Minitab, S-PLUS or Mathematica.
In Table 5 we have provided the limits for the case
of the Jeffreys prior for 7 ≤ n ≤ 30.
The endpoints of the Jeffreys prior interval are
the α/2 and 1−α/2 quantiles of the Betax+1/2 n−
x + 1/2 distribution. The psychological resistance
among some to using the interval is because of the
inability to compute the endpoints at ease without
software.
We provide two avenues to resolving this problem.
One is Table 5 at the end of the paper. The second
is a computable approximation to the limits of the
Jeffreys prior interval, one that is computable with
just a normal table. This approximation is obtained
after some algebra from the general approximation
to a Beta quantile given in page 945 in Abramowitz
and Stegun (1970).
The lower limit of the 1001 − α% Jeffreys prior
interval is approximately
x + 1/2
n + 1 + n − x + 1/2e2ω − 1 (9)
where
ω = κ
4pˆq/n ˆ + κ2 − 3/6n2
4pˆqˆ
+ 1/2 − ˆp ˆpqˆκ2 + 2 − 1/n
6n ˆpqˆ2
The upper limit may be approximated by the same
expression with κ replaced by −κ in ω. The simple
approximation given above is remarkably accurate.
Berry (1996, page 222) suggests using a simpler normal
approximation, but this will not be sufficiently
accurate unless npˆ1 − ˆp is rather large.
Jeffreys interval. Beta distributions are thestandard conjugate priors for binomial distributionsand it is quite common to use beta priors for inferenceon p (see Berger, 1985).Suppose X ∼ Binn p and suppose p has a priordistribution Beta a1 a2; then the posterior distributionof p is BetaX + a1 n − X + a2. Thus a1001 − α% equal-tailedBayesian interval is givenbyBα/2 X + a1 n − X + a2B1 − α/2 X + a1 n − X + a2where Bα m1 m2 denotes the α quantile of aBeta m1 m2 distribution.The well-known Jeffreys prior and the uniformprior are each a beta distribution. The noninformativeJeffreys prior is of particular interest to us.Historically, Bayes procedures under noninformativepriors have a track record of good frequentistproperties; see Wasserman (1991). In this problemthe Jeffreys prior is Beta1/2 1/2 which has thedensity functionfp = π−1p−1/21 − p−1/2 The 1001 − α% equal-tailed Jeffreys prior intervalis defined as(6) CIJ = LJx UJxwhere LJ0 = 0 UJn = 1 and otherwise(7) LJx = Bα/2 X + 1/2 n − X + 1/2(8) UJx = B1 − α/2 X + 1/2 n − X + 1/2 The interval is formed by taking the central 1 − αposterior probability interval. This leaves α/2 posteriorprobability in each omitted tail. The exceptionis for x = 0n where the lower (upper) limits aremodified to avoid the undesirable result that thecoverage probability Cp n → 0 as p → 0 or 1.The actual endpoints of the interval need to benumerically computed. This is very easy to do usingsoftwares such as Minitab, S-PLUS or Mathematica.In Table 5 we have provided the limits for the caseof the Jeffreys prior for 7 ≤ n ≤ 30.The endpoints of the Jeffreys prior interval arethe α/2 and 1−α/2 quantiles of the Betax+1/2 n−x + 1/2 distribution. The psychological resistanceamong some to using the interval is because of theinability to compute the endpoints at ease withoutsoftware.We provide two avenues to resolving this problem.One is Table 5 at the end of the paper. The secondis a computable approximation to the limits of theJeffreys prior interval, one that is computable withjust a normal table. This approximation is obtainedafter some algebra from the general approximationto a Beta quantile given in page 945 in Abramowitzand Stegun (1970).The lower limit of the 1001 − α% Jeffreys priorinterval is approximatelyx + 1/2n + 1 + n − x + 1/2e2ω − 1 (9) whereω = κ4pˆq/n ˆ + κ2 − 3/6n24pˆqˆ+ 1/2 − ˆp ˆpqˆκ2 + 2 − 1/n6n ˆpqˆ2 The upper limit may be approximated by the sameexpression with κ replaced by −κ in ω. The simpleapproximation given above is remarkably accurate.Berry (1996, page 222) suggests using a simpler normalapproximation, but this will not be sufficientlyaccurate unless npˆ1 − ˆp is rather large.
การแปล กรุณารอสักครู่..
