The Control of the False Discovery Rate in Multiple Testing under Dependency

Author(s): Yoav Benjamini and Daniel Yekutieli

Source: The Annals of Statistics, Vol. 29, No. 4 (Aug., 2001), pp. 1165-1188
Published by: Institute of Mathematical Statistics
Stable URL: https://www.jstor.org/stable/2674075
Accessed: 30-05-2025 09:45 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Institute of Mathematical Statistics is collaborating with JSTOR to digitize, preserve and

extend access to The Annals of Statistics

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
The Annals of Statistics
2001, Vol. 29, No. 4, 1165-1188

THE CONTROL OF THE FALSE DISCOVERY RATE IN

MULTIPLE TESTING UNDER DEPENDENCY

BY YOAV BENJAMINI1 AND DANIEL YEKUTIELI2

Tel Aviv University

Benjamini and Hochberg suggest that the false discovery rate may be
the appropriate error rate to control in many applied multiple testing prob-
lems. A simple procedure was given there as an FDR controlling procedure
for independent test statistics and was shown to be much more powerful
than comparable procedures which control the traditional familywise error
rate. We prove that this same procedure also controls the false discovery
rate when the test statistics have positive regression dependency on each of
the test statistics corresponding to the true null hypotheses. This condition
for positive dependency is general enough to cover many problems of prac-
tical interest, including the comparisons of many treatments with a single
control, multivariate normal test statistics with positive correlation matrix
and multivariate t. Furthermore, the test statistics may be discrete, and
the tested hypotheses composite without posing special difficulties. For all
other forms of dependency, a simple conservative modification of the proce-
dure controls the false discovery rate. Thus the range of problems for which
a procedure with proven FDR control can be offered is greatly increased.

1. Introduction.

1.1. Simultaneous hypotheses testing. The control of the increased type I

error when testing simultaneously a family of hypotheses is a central issue in
the area of multiple comparisons. Rarely are we interested only in whether
all hypotheses are jointly true or not, which is the test of the intersection null
hypothesis. In most applications, we infer about the individual hypotheses,
realizing that some of the tested hypotheses are usually true-we hope not
all-and some are not. We wish to decide which ones are not true, indicating
(statistical) discoveries. An important such problem is that of multiple end-
points in a clinical trial: a new treatment is compared with an existing one in
terms of a large number of potential benefits (endpoints).

EXAMPLE 1.1 (Multiple endpoints in clinical trials). As a typical example,

consider the double-blind controlled trial of oral clodronate in patients with
bone metastases from breast cancer, reported in Paterson, Powles, Kanis,
McCloskey, Hanson and Ashley (1993). Eighteen endpoints were compared

Received February 1998; revised April 2001.

1Supported by FIRST foundation of the Israeli Academy of Sciences and Humanities.
2This article is a part of the author's Ph.D. dissertation at Tel Aviv University, under the
guidance of Yoav Benjamini.
AMS 2000 subject classifications. 62J15, 62G30, 47N30.
Key words and phrases. Multiple comparisons procedures, FDR, Simes' equality, Hochberg's
procedure, MTP2 densities, positive regression dependency, unidimensional latent variables, dis-
crete test statistics, multiple endpoints many-to-one comparisons, comparisons with control.
1165

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1166 Y. BENJAMINI AND D. YEKUTIELI
between the treatment and the control groups. These endpoints included,
among others, the number of patients developing hypercalcemia, the num-
ber of episodes, the time the episodes first appeared, number of fractures
and morbidity. As is clear from the condensed information in the abstract,
the researchers were interested in all 18 particular potential benefits of the
treatment.

The traditional concern in such multiple hypotheses testing problems has

been about controlling the probability of erroneously rejecting even one of the
true null hypotheses, the familywise error-rate (FWE). Books by Hochberg
and Tamhane (1987), Westfall and Young (1993), Hsu (1996) and the review
by Tamhane (1996) all reflect this tradition. The control of the FWE at some
level a requires each of the individual m tests to be conducted at lower levels,
as in the Bonferroni procedure where a is divided by the number of tests
performed.
The Bonferroni procedure is just an example, as more powerful FWE con-
trolling procedures are currently available for many multiple testing problems.
Many of the newer procedures are as flexible as the Bonferroni, making use of
the p-values only, and a common thread is their stepwise nature (see recent
reviews by Tamhane (1996), Shaffer (1995) and Hsu (1996)). Still, the power
to detect a specific hypothesis while controlling the FWE is greatly reduced
when the number of hypotheses in the family increases, the newer procedures
notwithstanding. The incurred loss of power even in medium size problems
has led many practitioners to neglect multiplicity control altogether.

EXAMPLE 1.1 (Continued). Paterson et al. (1993) summarize their results

in the abstract as follows:

In patients who received clodronate, there was a significant reduction

compared with placebo in the total number of hypercalcemic episodes
(28 v 52; p < .01), in the number of terminal hypercalcemic episodes (7
v 17; p < .05), in the incidence of vertebral fractures (84 v 124 per 100
patient-years; p < .025), and in the rate of vertebral deformity (168 v
252 per 100 patient-years; p < .001)...

All six p-values less than 0.05 are reported as significant findings. No
adjustment for multiplicity was tried nor even a concern voiced.

While almost mandatory in psychological research, most medical journals

do not require the analysis of the multiplicity effect on the statistical conclu-
sions, a notable exception being the leading New England Journal of Medicine.
In genetics research, the need for multiplicity control has been recognized as
one of the fundamental questions, especially since entire genome scans are
now common [see Lander and Botstein (1989), Barinaga (1994), Lander and
Kruglyak (1995), Weller, Song, Heyen, Lewin and Ron (1998)]. The appropri-
ate balance between lack of type I error control and low power ["the choice

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1167

between Scylla and Charybdis" in Lander and Kruglyak (1995)] has been
heavily debated.

1.2. The false discovery rate. The false discovery rate (FDR), suggested by
Benjamini and Hochberg (1995) is a new and different point of view for how
the errors in multiple testing could be considered. The FDR is the expected
proportion of erroneous rejections among all rejections. If all tested hypotheses
are true, controlling the FDR controls the traditional FWE. But when many
of the tested hypotheses are rejected, indicating that many hypotheses are
not true, the error from a single erroneous rejection is not always as crucial
for drawing conclusions from the family tested, and the proportion of errors
is controlled instead. Thus we are ready to bear with more errors when many
hypotheses are rejected, but with less when fewer are rejected. (This frequen-
tist goal has a Bayesian flavor.) In many applied problems it has been argued
that the control of the FDR at some specified level is the more appropriate
response to the multiplicity concern: examples are given in Section 2.1 and
discussed in Section 4.
The practical difference between the two approaches is neither trivial nor
small and the larger the problem the more dramatic the difference is. Let us
demonstrate this point by comparing two specific procedures, as applied to
Example 1.1. To fix notation, let us assume that of the m hypotheses tested
{H?, Ho, ..., Ho }, mo are true null hypotheses, the number and identity of
which are unknown. The other m - mo hypotheses are false. Denote the cor-
responding random vector of test statistics {X1, X2, ..., Xm}, and the corre-
sponding p-values (observed significance levels) by {P1, P2,..., Pm } where
Pi = 1- FHo(Xi).
Benjamini and Hochberg (1995) showed that when the test statistics are
independent the following procedure controls the FDR at level q mo/m < q.

THE BENJAMINI HOCHBERG PROCEDURE. Let P(1) < P(2) < < P(m) be the
ordered observed p-values. Define

(1) k = max i: p(i) < q

and reject H) ... H(k). If no such i exists, reject no hypothesis.
In the case that all tested hypotheses are true, that is, when mo = m, this
theorem reduces to Simes' global test of the intersection hypothesis proved
first by Seeger (1968) and then independently by Simes (1986). However, when
mo < m the procedure does not control the FWE. To achieve FWE control,
Hochberg (1988) constructed a procedure from the global test, which has the
same stepwise structure but each P(i) is compared to m-q1 instead of i.
The constants for the two procedures are the same at i = 1 and i = m but
elsewhere the FDR controlling constants are larger.

EXAMPLE 1.1 (Continued). Compare the two procedures conducted at the

0.05 level in the multiple endpoint example. Hochberg's FWE controlling pro-

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1168 Y BENJAMINI AND D. YEKUTIELI
cedure rejects the two hypotheses with p-values less than 0.001, just as the
Bonferroni procedure does. The FDR controlling procedure rejects the four
hypotheses with p-values less than 0.01. In this study the ninth p-value is
compared with 0.005 if FWE control is required, with 0.025 if FDR control is
desired.

More details about the concept and procedures, other connections and his-
torical references are discussed in Section 2.2.

1.3. The problem. When trying to use the FDR approach in practice,
dependent test statistics are encountered more often than independent ones,
the multiple endpoints example of the above being a case in point. A simulation
study by Benjamini, Hochberg and Kling (1997) showed that the same proce-
dure controls the FDR for equally positively correlated normally distributed
(possibly Studentized) test statistics. The study also showed, as demonstrated
above, that the gain in power is large. In the current paper we prove that the
procedure controls the FDR in families with positively dependent test statis-
tics (including the case investigated in the mentioned simulation study). In
other cases of dependency, we prove that the procedure can still be easily modi-
fied to control the FDR, although the resulting procedure is more conservative.
Since we prove the theorem for the case when not all tested hypotheses
are true, the structure of the dependency assumed may be different for the
set of the true hypotheses and for the false. We shall obviously assume that
at least one of the hypotheses is true, otherwise the FDR is trivially 0. The
following property, which we call positive regression dependency on each one
from a subset Io, or PRDS on Io, captures the positive dependency structure
for which our main result holds. Recall that a set D is called increasing if
x E D and y > x, implying that y E D as well.

PROPERTY PRDS. For any increasing set D, and for each i E Io, P(X E D
Xi = x) is nondecreasing in x.

The PRDS property is a relaxed form of the positive regression dependency

property. The latter means that for any increasing set D, P(X E D I X1 =
x, ... . Xi = xi) is nondecreasing in (x1, ..., xi) [Sarkar (1969)]. In PRDS the
conditioning is on one variable only, each time, and required to hold only for
a subset of the variables. If X is MTP2, X is positive regression dependent,
and therefore also PRDS over any subset (details in Section 2.3), a property
we shall simply refer to as PRDS.

1.4. The results. We are now able to state our main theorems.

THEOREM 1.2. If the joint distribution of the test statistics is PRDS on the
subset of test statistics corresponding to true null hypotheses, the Benjamini
Hochberg procedure controls the FDR at level less than or equal to m? q.

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1169

In Section 2 we discuss in more detail the FDR criterion, the historical

background of the procedure and available results and review the relevant
notions of positive dependency. This section can be consulted as needed. In
Section 3 we outline some important problems where it is natural to assume
that the conditions of Theorem 1.2 hold. In Section 4 we prove the theorem.
In the course of the proof we provide an explicit expression for the FDR, from
which many more new properties can be derived, both for the independent and
the dependent cases. Thus issues such as discrete test statistics, composite
null hypotheses, general step-up procedures and general dependency can be
addressed. This is done in Section 5. In particular we prove there the following
theorem.

THEOREM 1.3. When the Benjamini Hochberg procedure is conducted with

q/(L7=1 1) taking the place of q in (1), it always controls the FDR at level less
than or equal to 'O q.

As can be seen from the above summary, the results of this article greatly
increase the range of problems for which a powerful procedure with proven
FDR control can be offered.

2. Background.

2.1. The FDR criterion. Formally, as in Benjamini and Hochberg (1995),

let V denote the number of true null hypotheses rejected and R the total num-
ber of hypotheses rejected, and let Q be the unobservable random quotient,

Q_ |V/R, if R,>O~
Q /R, otherwise.
Then the FDR is simply E(Q). Their approach calls for controlling the FDR
at a desired level q, while maximizing E(R).
If all null hypotheses are true (the intersection null hypothesis holds) the
FDR is the same as the probability of making even one error. Thus controlling
the FDR controls the latter, and q is maybe chosen at the conventional levels
for a. Otherwise, when some of the hypotheses are true and some are false, the
FDR is smaller [Benjamini and Hochberg (1995)]. The control of FDR assumes
that when many of the tested hypotheses are rejected it may be preferable to
control the proportion of errors rather than the probability of making even
one error.
The FDR criterion, and the step-up procedure that controls it, have been
used successfully in some very large problems: thresholding of wavelets coeffi-
cients [Abramovich and Benjamini (1996)], studying weather maps [Yekutieli
and Benjamini (1999)] and multiple trait location in genetics [Weller et al.
(1998)], among others. Another attractive feature of the FDR criterion is that
if it is controlled separately in several families at some level, then it is also
controlled at the same level at large (as long as the families are large enough,
and do not consist only of true null hypotheses).

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1170 Y. BENJAMINI AND D. YEKUTIELI
Although the FDR controlling procedure has been implemented in standard
computer packages (MULTPROC in SAS), one of its merits is the simplicity
with which it can be performed by succinct examination of the ordered list
of p-values from the largest to the smallest, and comparing each P(i) to i
times q/m stopping at the first time the former is smaller than the latter and
rejecting all hypotheses with smaller p-values. Rough arithmetic is usually
enough.

2.2. Positive dependency. Lehmann (1996) first suggested a concept for

bivariate positive dependency, which is very close to the above one and
amounts to being PRDS on every subset. Generalizing his concept from bivari-
ate distributions to the multivariate ones was done by Sarkar (1969). A mul-
tivariate distribution is said to have positive regression dependency if for any
increasing set D, P(X E D I X1 = xl, ..., Xi = xi) is nondecreasing in
(Xi, . * *, Xi).
A stricter condition, implying positive regression dependency, is multivari-
ate total positivity of order 2, denoted MTP2: X is MTP2 if for all x and y,
(2) f(x). f(y) < f(min(x, y)). f(max(x, y)),
where f is either the joint density or the joint probability function, and the
minimum and maximum are evaluated componentwise. While being a strong
notion of dependency, MTP2 is widely used, as this property is easier to show.
Positive regression dependence implies in turn that X is positive associated,
in the sense that for any two functions f and g, which are both increasing (or
both decreasing) in each of the coordinates, cov(f(X)g(X)) > 0.
PRDS has two properties in which it is different from the above concept.
First, monotonicity is required after conditioning only on one variable at a
time. Second, the conditioning is done only on any one from a subset of the
variables. Thus if X is MTP2, or if it is positive regression dependent, then
it is obviously positive regression dependent on each one from any subset.
Nevertheless, PRDS and positive association do not imply one another, and
the difference is of some importance. For example, a multivariate normal dis-
tribution is positively associated iff all correlations are nonnegative. Not all
correlations need be nonnegative for the PRDS property to hold (see Section
3.1, Case 1 below). On the other hand, a bivariate distribution may be posi-
tively associated, yet not positive regression dependent [Lehmann (1966)], and
therefore also not PRDS on any subset. A stricter notion of positive associa-
tion, Rosenbaum's (1984) conditional (positive) association, is enough to imply
PRDS: X is conditionally associated, if for any partition (X1, X2) of X, and any
function h(X1), X2 given h(Xl) is positively associated.
It is important to note that all of the above properties, including PRDS,
remain invariant to taking comonotone transformations in each of the coor-
dinates [Eaton (1986)]. Note also that D is increasing iff D is decreasing, so
the PRDS property can equivalently be expressed by requiring that for any
decreasing set C, and for each i E Io, P(X E C I Xi = x) is nonincreasing
in x. Therefore, whenever the joint distribution of the test statistics is PRDS

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1171

on some Io so is the joint distribution of the corresponding p-values, be they

right-tailed or left-tailed. Background on these concepts is clearly presented
in Eaton (1986), supplemented by Holland and Rosenbaum (1986).

2.3. Historical background and related results. The FDR controlling mul-
tiple testing procedure [Benjamini and Hochberg (1995)], given by (1), is a
step-up procedure that involves a linear set of constants on the p-value scale
(step-up in terms of test statistics, not p-values). The FDR controlling pro-
cedure is related to the global test for the intersection hypothesis, which is
defined in terms of the same set of constants: reject the single intersection
hypothesis if there exist an i s.t. P(i) < ma. Simes (1986) showed that when
the test statistics are continuous and independent, and all hypotheses are
true, the level of the test is a. The equality is referred to as Simes' equality,
and the test has been known in recent years as Simes' global test. However
the result had already been proved by Seeger (1968) [Shaffer (1995) brought
this forgotten reference to the current literature.] See Sen (1999a, b) for an
even earlier, though indirect, reference.
Simes (1986) also suggested the procedure given by (1) as an informal mul-
tiple testing procedure, and so did Elkund, some 20 years earlier [Seeger
(1968)]. The distinction between a global test and a multiple testing proce-
dure is important. If the single intersection hypothesis is rejected by a global
test, one cannot further point at the individual hypotheses which are false.
When some hypotheses are true while other are false (i.e., when mo < m),
Seeger (1968) showed, referring to Elkund, and Hommel (1988) showed, refer-
ring to Simes, that the multiple testing procedure does not necessarily control
the FWE at the desired level. Therefore, from the perspective of FWE control,
it should not be used as a multiple testing procedure. Other multiple testing
procedures that control the FWE have been derived from the Seeger-Simes
equality, for example, by Hochberg (1988) and Hommel (1988).
Interest in the performance of the global test when the test statistics are
dependent started with Simes (1986), who investigated whether the procedure
is conservative under some dependency structures, using simulations. On the
negative side, it has been established by Hommel (1988) that the FWE can
get as high as a (1 + 1/2 + +.. ? 1/m). The joint distribution for which this
upper bound is achieved is quite bizarre, and rarely encountered in practice.
But even with tamed distributions, the global test does not always control
the FWE at level a. For example, when two test statistics are normally dis-
tributed with negative correlation the FWE is greater than a, even though the
difference is very small for conventional levels [Hochberg and Rom (1995)].
On the other hand, extensive simulation studies had shown that for posi-
tive dependent test statistics, the test is generally conservative. These results
were followed by efforts to extend theoretically the scope of conservativeness,
starting with Hochberg and Rom (1995). These efforts have been reviewed in
the most recent addition to this line of research by Sarkar (1998). An exten-
sive discussion with many references can be found in Hochberg and Hommel
(1998).

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1172 Y. BENJAMINI AND D. YEKUTIELI
Directly relevant to our work are the two strongest results for positive
dependent test statistics: Chang, Rom and Sarkar (1996) proved the conser-
vativeness for multivariate distributions with MTP2 densities. The condition
for positive dependency is weaker in the first but the proof applies to bivariate
distributions only. Theorem 1.2, when applied to the limited situation where
all null hypotheses are true, generalizes the result of Chang, Rom and Sarkar
(1996) to multivariate distributions. Although the final result is somewhat
stronger than that of Sarkar (1998), the generalization is hardly of impor-
tance for the limited case in which all tested hypotheses are true. The full
strength of Theorem 1.2 is in the situation when some hypotheses may be
true and some may be false, where the full strength of a multiple testing pro-
cedure is needed. For this situation the results of Section 2.1 for independent
test statistics are the only ones available.

3. Applications. In the first part of this section we establish the PRDS

property for some commonly encountered distributions. Recall the sets of vari-
ables we have: test statistics for which the tested hypotheses are true and test
statistics for which they are false. We are inclined to assume less about the
joint distribution of the latter, as will be reflected in some of the following
results. In the second part we review some multiple hypotheses testing prob-
lems where controlling the FDR is desirable, and where applying Theorem 1.2
shows that using the procedure is a valid way to control it. We emphasize the
normal distribution and its related distributions in the first part. For many
of the examples in the second part, using normal distribution assumptions
for the test statistics is only a partial answer, as methods which are based
on other distributions for the test statistics are sometimes needed (such as
nonparametric). These issues are beyond the scope of this study.

3.1. Distributions.

CASE 1 (Multivariate normal test statistics). Consider X - N(,t, 1) a vec-

tor of test statistics each testing the hypothesis Aui = 0 against the alternative
,ui > 0, for i = 1, ..., m. For i E Io, the set of true null hypotheses, t = 0.
Otherwise 1xi > 0.
Assume that for each i E Io, and for each j 0 i, tU > 0, then the distribu-
tion of X is PRDS over Io.

PROOF. For any i E Io, denote by X(i) the remaining m - 1 test statistics,
Aw is its mean vector, Y(i) i is the column of covariances of Xi with X(i), and
(i, i) is X after dropping the ith row and column.
The distribution of X) given Xi xi is N(,u(i), V(i)), where

1( ) X ) - X(i), i ) and ,ji) = I(i) + X(i), ilI( (Xi - A)

Thus if 1(i) i is positive, the conditional means increase in xi. Since the covari-
ance remains unchanged, the conditional distribution increases stochastically

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1173

as xi increases; that is, for any increasing functions f, if xi < x' then

(3) E(f(X(')) I Xi = xi) < E(f(X(')) I Xi = x).

Hence the PRDS over Io holds.

Note that the intercorrelations among the test statistics corresponding to

the false null hypotheses need not be nonnegative. The fact that less struc-
ture is imposed under the alternative hypotheses may be important in some
applications; see, for example, the multiple endpoints problem in the following
section.

CASE 2 (Latent variable models). In monotone latent variable models, the

distribution of X is assumed to be the marginal distribution of some (X, U),
where the components of X given U = u are (a) independent, and (b) stochas-
tically comonotone in u.
If, furthermore, U is univariate, X is said to have a unidimensional latent
variable distribution [Holland and Rosenbaum (1986)]. Holland and
Rosenbaum (1986) show that a unidimensional latent distribution is condi-
tionally positively associated. Therefore it is also PRDS on any subset.
It is interesting to note that the distributions for which Sarkar and Chang
(1997) prove their result are all unidimensional latent variable distributions.
For the multivariate latent variable model, if U is MTP2, and each Xi I
U = u is MTP2 in xi and u, then the distribution of X is MTP2 (called latent
MTP2.) See again Holland and Rosenbaum (1986), based on a lemma of Karlin
and Rinott (1980). While MTP2 is not enough to imply conditional positive
association, it is enough to assure PRDS over any subset.
We shall now generalize the unidimensional latent variable models, to dis-
tributions in which the conditional distribution of X given U is not
independent but PRDS on a subset Io. In this class of distributions the ran-
dom vector X is expressed as a monotone transformation of a PRDS ran-
dom vector Y and an independent latent variable U, the components of X are
Xi = g1(Yi, U).

LEMMA 3.1. If (a) Y is a continuous random vector, PRDS on a subset

Io; (b) U an independently distributed continuous random variable; (c) for j =
1 - m the components of X, Xj = gj(Yj, U) are strictly increasing continuous
functions of the coordinates Yj and of U; (d) for i E Io, U and Yi are PRDS
on Xi; then X is PRDS on Io.

The proof of this lemma is somewhat delicate and lengthy and is given in
the Appendix. Condition (d) of the lemma depends on both the transformation
gi and the distribution of Yi and U. In the following example condition (d) is
asserted via the stronger TP2 condition.

EXAMPLE 3.2. U0 and U1 are independent chi-square or inverse chi-square

random variables, W = U0 . U1. We show that Ui is PRDS on W by showing

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1174 Y. BENJAMINI AND D. YEKUTIELI
the TP2 property for each pair (Ui, W), i = 0, 1. Since for i = 0, 1,

f uj W(X1, X2) = 1/X1 * fui (XI) - f Uli_(X21Xl),

it is sufficient to assert that fUl-i(X2/X1) is TP2 in xl and x2. It is easy to

check that this property holds for both the chi-square and inverse chi-square
distributions.

COROLLARY 3.3. If Y is multivariate normal, IYI PRDS on the subset Io

for which 1Ai = 0 and S2 is an independently distributed X2, then IXj = IYI/S
is PRDS on Io.

PROOF. Using Example 3.2, setting UO - lY2 and U1 = 1/S2, condition

(d) holds so we can apply Lemma 3.1.

CASE 3 (Absolute values of multivariate normal and t). Y - N(Qi, E) and

consider two-sided tests: /ii 0 O against the alternative ,ui =A 0. Test statistics
are multivariate t, obtained by dividing IYI by an independent (pooled) chi-
square distributed estimator S > 0. According to Corollary 3.3 if JYJ is PRDS
over the set of true null hypotheses then IYI/S is also PRDS over the set of
true null hypotheses.
If E = I, the components of IYJ are independent and thus PRDS over any
subset. For E # I, IYI is known to be MTP2 under some conditions [see Karlin
and Rinott (1981)], but only when all ,ui = 0. This case was already covered
by Sarkar (1998) and is an uncommon example in which all null hypotheses
are true, hence the FDR equals the FWE.
Y can also contain a subset of dependent ,u = 0 components of the above
form and a subset of ,t 0 0 components, each component corresponding to
,u = 0 independent of all , :A 0 components; IYI is then PRDS over the subset
for which ,t = 0.

CASE 4 (Studentized multivariate normal). Consider now Y multivariate

normal as in Case 1, Studentized as in Case 3 by S. Because the direction
of monotonicity of Yi/S in S changes as the sign of Yi changes, Y/S is not
PRDS. Yet we will now show that if q, the level of the test, is less than 1/2,
the Benjamini Hochberg procedure applied to Y/S offers FDR control.
We will show this by introducing a new random vector S+(Y, S) defined as
follows: if Yj > 0 then S+(Yj, S) = Yj/S, otherwise S+(Y , S) = Y1. The
transformation S+(Y, S) is increasing in both Yj and in 1/S, which satis-
fies condition (c) in Lemma 3.1. Condition (d) of Lemma 3.1 is also kept, but
only for positive values of Yi, for which we can express S+(Yi, S) = lYiI/S.
According to Remark A.4 in the Appendix, S+(Y, S) is PRDS, but only when
the conditioning is on positive values of S+(Yi, S).
According to Remark 4.2, the PRDS condition must only hold for Pi E [0, q].
For q < 1/2 this means positive value of S+(Yi, S). Hence when applied to
S+(Y, S) procedure (1) controls the FDR.
Finally notice that since q < 1/2 all the critical values of procedure (1)
are positive, and for Y > 0, S+(Y, S) _ Y/S. Hence the outcome of applying

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1175

procedure (a) on Y/S is identical to the outcome of applying procedure (1)

on S+(Y, S), therefore procedure (1) will also control the FDR when applied
to Y/S.

3.2. Applied problems.

PROBLEM 1 [Subgroup (subset) analysis in the comparison of two treat-

ments]. When comparing a new treatment to a common one, it is usually of
interest to find subgroups for which the new treatment may prove to be better.
If there is no "pooling" across subgroups involved, then the test statistics are
independent. More typically, averages are compared within the subgroups, yet
a pooled estimator of the standard deviation Spooled is used. Hence we have
test statistics which are independent and approximately normal, conditionally
on Spooled- These (usually) one-sided correlated t-tests fall under Case 4, and
thus Theorem 1.2 applies.

PROBLEM 2 (Screening orthogonal contrasts in a balanced design). Con-

sider a balanced factorial experiment with m factorial combinations and n
repetitions per cell, which is performed for the purpose of screening many
potential factors for their possible effect on a quantity of interest. Such exper-
iments are common, for example, in industrial statistics when screening for
possible factors affecting quality characteristics, and in the pharmaceutical
industry when screening for potentially beneficial compounds. In the above
two, economic considerations make it clear that in identifying a set of hypothe-
ses for further research, allowing a controlled proportion of errors in the iden-
tified pool is desirable. In fact the chosen level for q may be higher than the
levels usually used for a. The distributional model is that of (usually) two-
sided correlated t-tests, which thus fall under Case 3.

PROBLEM 3 (Many-to-one comparisons in clinical trials). Differently

phrased this is the problem of comparing a few treatments with a single con-
trol, using one-sided tests. See the recent review by Tamhane and Dunnett
(1999) for the many approaches and procedures that control the FWE. If the
interest lies in recommending one of the tested treatments based solely on the
current experiment, FWE should be controlled. But if the conclusion is closer
in nature to the conclusion of Problem 2, the control of FDR is appropriate
[see detailed discussion in Benjamini, Hochberg and Kling (1993)].
In the normal model, Xi = (Yi - Y0)/cjS, Yi, i = 0, 1, ..., m independent
normal random variables, with variances ci o-2 which are known up to o-, S2, an
independent estimator such that S2/0-2 _ X2/v. (Yi - Yo)/ci is multivariate
normal with pij > 0, hence PRDS, thus according to Case 4, X is PRDS on the
set of true null hypotheses.

EXAMPLE 3.4. The study of uterine weights of mice reported by Steel and
Torrie (1980) and discussed in Westfall and Young (1993) comprised a com-
parison of six groups receiving different solutions to one control group. The

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1176 Y. BENJAMINI AND D. YEKUTIELI
lower-tailed p-values of the pooled variance t-statistics are 0.183, 0.101, 0.028,
0.012, 0.003, 0.002. Westfall and Young (1993) show that, using p-value resam-
pling and step-down testing, three hypotheses are rejected at FWE 0.05. Four
hypotheses are rejected when applying procedure (1) using FDR level
of 0.05.

PROBLEM 4 (Multiple endpoints in clinical trials). Multiple endpoints, that

is, the multiple outcomes according to which the therapeutic properties of one
treatment are compared with those of an established treatment, raises one
of the most serious multiplicity control problems in the design and analy-
sis of clinical trials. For a recent review, see Wassmer, Reitmer, Kieser and
Lehmacher (1998). Eighteen outcomes were studied in Example 1.1, but the
number may reach hundreds, so addressing this problem by controlling the
FWE is overwhelmingly conservative. A common remedy is to specify very
few primary endpoints on which the conclusion will be based and give a lesser
standing to the conclusions from the other secondary endpoints, for which
FWE is not controlled. However, it is not uncommon to find the advocated
features of a new treatment to come mostly from the secondary endpoints.
The FDR approach is very natural for this problem, and the emphasise on
primary endpoints is no longer essential [but feasible as in Benjamini and
Hochberg (1997)].
The test statistics of the different endpoints are usually dependent. Their
dependency is in most cases neither constant nor known, and stems both
from correlated treatment effect (for nonnull treatment effects) and a latent
individual component affecting the value of all endpoints of the same person.
The individual component introduces a latent positive dependence between all
test statistics. Thus test statistics of null hypotheses are positively correlated
with all other test statistics. Treatment effect may introduce negative correla-
tion between the affected endpoints, which may dominate the latent positive
dependency. Thus we want to allow those endpoints which are affected by the
treatment to have whatever dependence structure occurs among themselves.
Then, using the results of Cases 1, 2 and 4 above, Theorem 1.2 applies for
the one-sided tests, be they normal tests or t-tests. The situation with two-
sided tests is more complicated, as Case 3 requires a stronger assumption.

EXAMPLE 3.5 (Low lead levels and IQ). Needleman, Gunnoe, Leviton,
Reed, Presie, Maher and Barret (1979) studied the neuropsychologic effects
of unidentified childhood exposure to lead by comparing various psycholog-
ical and classroom performances between two groups of children differing
in the lead level observed in their shed teeth. While there is no doubt that
high levels of lead are harmful, Needleman's findings regarding exposure to
low lead levels, especially because of their contribution to the Environmen-
tal Protection Agency's review of lead exposure standards, are controversial.
Needleman's study was attacked on the ground of methodological flaws; for
details see Westfall and Young (1993). One of the methodological flaws pointed
out is control of multiplicity. Needleman et al. (1979) present three families of

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1177
TABLE 1

p-values FWE FDR

(omitting sum Rej. # of Rej. # of
Family score p-values) thrshld. rej. thrshld. rej.
Teacher's behavioral 0.003 0.05 0.05 0.14 0.005 3 0.02 5
ratings 0.08 0.01 0.04 0.01
0.05 0.003 0.003
Score of Wechsler 0.04 0.05 0.02 0.49 0.004 0 0.004 0
Intelligence Scale 0.08 0.36 0.03 0.38
for Children (revised) 0.15 0.90 0.37 0.54
Verbal processing 0.002 0.03 0.07 0.37 0.004 3 0.016 4
and reaction times 0.90 0.42 0.05 0.04

The three families 0.001 2 0.012 9

0.32 0.001 0.0010.01

jointly

endpoints, and comment on the results of separate multiplicity adjustments

within each family as summarized in Table 1 (under the FWE heading).

The critics argue that multiplicity should be controlled for all families
jointly. Using Hochberg's method at 0.05 level, correcting within each fam-
ily, six hypotheses are rejected. Correcting for all 35 responses, lead is found
to have an adverse effect in only two out of 35 endpoints.
Applying procedure (1) at 0.05 FDR level, the attack on Needleman findings
on grounds of inadequate multiplicity control is unjustified; whether analyzed
jointly or each family separately, lead was found to have an adverse effect in
more than a quarter of the endpoints.

4. Proof of theorem. For ease of exposition let us denote the set of con-
stants in (1), which define the procedure, by

(4) qi=-q, i = 1,2,...,Im. m

Let A s denote the event that the Benjamini Hochberg procedure rejects
exactly v true and s false null hypotheses. The FDR is then
ml mO

s=O v=l v + S

In the following lemma, Pr(Av ,) is expressed as an average.

LEMMA 4.1.

(6) Pr(A, s) = - E Pr({Pi ' qv+s} n Av, s)

1 mO

i=1

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1178 Y. BENJAMINI AND D. YEKUTIELI
PROOF. For a fixed v and s, let w denote a subset of {1... mO} of size v,
and AO s the event in A that the v true null hypotheses rejected are w. Note
that Pr{Pi < qv+s n A, 5} equals Pr{A, J} if i E wt), and is otherwise 0.

i-=1 i=l (0
LPr({Pi < q,e,} n Av, s) Pr ({Pi < qv+sl n AO

(7) ( i1 mO

-PzLLPr({Pi < qv+sl nA")

mO

E>j I(i e ) Pr{A, s}

=Ev Pr (A,S) = vPr{Av,s}.

Combining equation (5) with Lemma 4.1, the FDR is

ml mO M ?O1
(8) S=O V V-+S| V Pr( Pi ' qV+S} n Av s)}
(8)s0v1j 0 i1 mli is A 1 1
i=O S=O =1U?S

Now that the dependency of the expectation on v is only through Av s; we

reconstruct AU s from events that depend on i and k = v + s only, so the FDR
may be expressed similarly.
For i = 1 ... mo, let P(i) be the remaining m - 1 p-values after dropping
Pi. Let C(i) denote the event in which if Pi is rejected then v - 1 true null
hypotheses and s false null hypotheses are rejected alongside with it. That
is, C()s is the projection of {Pi < qvlJ n AU s onto the range of P(i), and
expanded again by cross multiplying with the range of Pi. Thus we have

(9) {Pi < qv+sj n Av, - {Pi < q0f8} An Cv,$s.

Denote by C(t) - U{C('): v + s = k}. For each i the C(k) are disjoint, so the
FDR can be expressed as

(10) E(Q) = L L k Pr (Pi < qk n C()

i=1k=

where the expression no longer depends on v and s, as desired.

In the last part of the proof we construct an expanding series of increasing
sets, on which we use the PRDS property to bound the inner sum in (8) by
q/m. For this purpose, define D(i) - U{C('): i < k} for k =1 m. D(i)

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1179

can also be described using the ordered set of the p-values in the range of
p(i) () < < P( 1), in the following way:
P(I) P(,-i) i) i)
(11) Dk = {p: qk+l < P(k)' qk+2 < P(k+l) ... <P(m-1)J
for k = 1... m - 1, and D(i) is simply the entire space. Expressing D(i) as
I ~~~~~~~~~~ ~~k
above, it becomes clear that for each k, D(i) is a nondecreasing set.
We now shall make use of the PRDS property, which states that for p < p',

(12) Pr(D I Pi = p) < Pr(D I Pi = p').

Following Lehmann (1996), it is easy to see that for j < I since qj < ql,

(13) Pr(D I Pi < qj) < Pr(D I Pi < ql),

for any nondecreasing set D, or equivalently,

Pr ({Pi < qk} n D5k) Pr ({Pi < qk+i} n D(k)

Pr(Pi < qk) Pr(Pi < qk+l)
j?1 j j+1
Invoking (14) together with the fact that D( = D(i) U CM yields for all
k < m - 1,

+ kI
Pr ({Pi < qk} n Dkk)+ Pr ({Pi < qk+1} n C(k)
Pr(Pi < qk) Pr(Pi < qk+l)
(15) < Pr ({Pi < qk+1} n Dk ) Pr ({Pi < qk+i} n C(5)J1)
Pr(Pi < qk+l) Pr(Pi < qk+l)
Pr ({ Pi < qk+l} n Dt1)
Pr (Pi < qk+l)

Now, start by noting that C1 = D1, and repeatedly use the above inequality
for i = 1, ..., m - 1, to fold the sum on the left into a single expression,

Pr({Pi < qk} n C(k)) Pr({Pi < qm} n Dm )=

(16) k=: Pr(Pi <~ qk) Pr(Pi < q.) =1
where the last equality follows because Dm) is the entire space.
Going back to expression (10) for the FDR,
mO m 1
E(Q) = ELk Pr ({Pi < q} n C)
i=1 k=k

)mO m q Pr ({Pi < qk} n Ck)

i=1 k=1m Pr(Pi < qk)

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1180 Y. BENJAMINI AND D. YEKUTIELI
because Pr(Pi < qk) < qk = mq under the null hypothesis (with equality for
continuous test statistics where each Pi is uniform), so finally, invoking (16),

(18) q m m Pr({Pi q} Ck)) < mOq

REMARK 4.2. Note that PRDS is a sufficient but not a necessary condition.
In particular the PRDS property need not hold for all monotone sets D and
all values of pi. According to inequality (12), it is enough that they hold for
monotone sets of the form of (11) and Pi E [0, q].

This remark is used to establish that Theorem 1.2 holds for one-sided mul-
tivariate t and q < 1/2, even though the distribution is not PRDS.

5. Generalizations and further results. If the test statistics are jointly

independent, the FDR as expressed in (10) is
MO M 1 k'
E(Q)= EL-Pr PC
i=1 k=1k ?}c5 )

(19) mo m 1 k P 1:Ek
(Pi < qmJ
Pr k
(C())
i=1 k=1

(20) P C') q i =m k=1 m

which yields an alternative (and possibly simpler) proof of the result in
Benjamini and Hochberg (1995). Moreover, the proof there depends critically
on the assumption that the P-values are uniformly distributed under the null
hypotheses, and therefore do not apply to discrete test statistics. However, for
discrete test statistics, we have that

(21) Pr(Pi <I) <q, i = 1,2, ...,mO.

Therefore, when passing from (19) to (20), we need only change the equality
to inequality in order to complete the proof of the following theorem.

THEOREM 5.1. For independent test statistics, the Benjamini Hochberg pro-
cedure controls the FDR at level less or equal to m? q. If the test statistics are
also continuous, the FDR is exactly m? q.

The argument leading to the above theorem used only the fact that for
discrete test statistics the tail probabilities are smaller. Thus, in a similar
way, it follows that the FDR is controlled when the procedure is used for
testing composite null hypotheses, as in one-sided tests.

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1181

THEOREM 5.2. For independent one-sided test statistics, if the distributions

in each of the composite null hypothesis are stochastically smaller than the null
distribution under which each p-value is computed, the Benjamini Hochberg
procedure controls the FDR at level less or equal to 'O q.

The surprising part of Theorem 5.1 is that equality holds no matter what
the distributions of the test statistics corresponding to the false null hypothe-
ses are. The following theorem shows that this is a unique property of the
step-up procedure which uses the constants { k q}. More generally, we can
define step-up procedures SU(&), using any other monotone series of con-
stants a1 < a,1 < < am_ let k = max{i: P(i) < ai}, and if such k exists
reject H(1)... H(k).

THEOREM 5.3. Testing m hypotheses with SU(ot), assume that the distri-
bution of the P-values, P = (PO, P1) is jointly independent.
(i) If the ratio ak/k is increasing in k, as the distribution of P1 increases
stochastically the FDR decreases.
(ii) If the ratio ak/k is decreasing in k, as the distribution of P1 increases
stochastically the FDR increases.

PROOF. Given the set of critical values a for k = 1, ..., m we define the
following sets:

(22) Ck(a) = P(i): P(k-1) < ak,., P(k) > ak+?1 (mr-) >1amj
Thus if p(i) E Ck(a) and Pi < ak then Ho is rejected along with k - 1 other
hypotheses, but if Pi > ak, Ho is not rejected. Notice that sets Ck(et) are
ordered. If p(i) E Ck(oa) and p(i) < pi(i), then all ordered coordinates of p,(i)
are greater or equal to corresponding coordinates of p(i). Therefore for j =
... m -1, P() > aj, thus p,(i) E Cl(a) for some 1 < k.
Next we define the function fa, fa: [0, l]m-1 f,

(23) f,(P(()) = ak/k for p(i) E Ck(0).

The FDR of all step-up procedures can be expressed similarly to expression
(10). Start deriving Lemma 4.1 by substituting ak in place of ak/m throughout
the proof. Then, denoting the FDR of SU(ox) by E(Q(ot)), we use the indepen-
dence of the test statistics to get
mO m 1
(24) E(Q( M)) = I < Pr({P <al} n P(i) E Ck(a))
i=1 k=1

mO m

(25) = m m Pr(Pi < aka)Pr (P(') E Ck(0)

mO m mO
i=1 k=1

(26) E aE ?/k Pr (P(i) E Ck()) =k Ep(iofa.

i=1 k=1 i=1
This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1182 Y. BENJAMINI AND D. YEKUTIELI
Note that the distribution of the test statistics corresponding to the mo
true null hypotheses is fully specified as U[O, 1]. If ak/k increases in k, the
function f, is a decreasing function. Stochastic increase in the distribution
of p(i) is characterized by the decrease of the expectation of all decreasing
functions, in particular a decrease in all the summands of the right side of (26).
Thus if P1 increases stochastically, the FDR decreases. If ak/k decreases in k,
the function fa is an increasing function. Thus if P1 increases stochastically
the FDR increases. (The case where ak/k is constant has been covered by
Theorem 5.1) 0

These more general step-up procedures are especially important in partic-

ular settings, where the structure of dependency can be precisely specified. In
such a case a specific set of constants can be used for designing a step-up pro-
cedure which exactly achieves the desired FDR at the specified distribution.
Troendle (1996) took this route, calculating a monotone series of constants,
which upon being used in the above fashion, control the FDR for normally
distributed test statistics which are equally and positively correlated. His cal-
culations were done under the unproven assertion that when the nonzero
means are set at infinity the FDR is maximized. In order to use Theorem
5.3 for that purpose it should be generalized first to hold under some joint
distribution other than independent, say PRDS. We do not have yet such a
result.
An important question that remains to be answered is the scope of problems
for which the two-sided tests retain the same level of control. Another impor-
tant open question is whether the same procedure controls the FDR when
testing pairwise comparisons of normal means, either Studentized or not.
Simulation studies, by Williams, Jones and Tukey (1999) and by Benjamini,
Hochberg and Kling (1993), and some limited calculations in the latter, show
that this is the case. It is known that the distribution of the test statistics is
not MTP2. The PRDS condition does not hold as well.
When facing such problems, it is always comforting to have a fallback pro-
cedure. The available FWE controlling procedure can be modified by working
at level a/ E U1 4, and it will then control the FWE at level a for any joint dis-
tribution of the test statistics-as long as the hypotheses are all true [Hommel
(1988)]. Similarly, Theorem 1.3 establishes that the same modification of the
procedure controls the FDR at the desired level, for any joint distribution of
the test statistics.

PROOF OF THEOREM 1.3. For simplicity of the exposition we shall use q in

(1), and show that the FDR is increased by no more than E71 1-
Denote Pikj = Pr({PQ E [(Pi l)q, q]} n C(5)). Note that,

m27) ,Pijk Pr (IPi E m k

k=1 - k=
(27) E: pjk =P Pi q-q n k

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1183

Returning to expression (10), the FDR can be expressed as

mO m 1 k mO m m 1
(28) E(Q) = E pijk = E PiJk
i=1 k=1 j=1 i=1 j=1 k=j
mO mm m MO m 1 m m lq
(29) ~~<i=1-Pi,k < P,7ik =mO E *C
j=1 k=j I i=1 j=1 k=1 j= J m
Obviously, as the main thrust of this paper shows, the adjustment by
log(m) + 2 is very often unneeded, and yields too conservative a
procedure. Still, even if only a small proportion of the tested hypotheses are
detected as not true [approximately log(m)/m], the procedure is more power-
ful than the comparable FWE controlling procedure of Holm (1979). The ratio
of the defining constants can get as high as (m + 1)/4log(m) in favor of the
FDR controlling procedure, so its advantage can get very large.
It should be noted that throughout all results of this work, the procedure
controls the FDR at a level too low by a factor of mo/m. Loosely speaking, the
procedure actually controls the false discovery likelihood ratio,

(30) E (m < q.
Other procedures, which get closer to controlling the FDR at the desired level,
have been offered for independent test statistics in Benjamini and Hochberg
(2000), and in Benjamini and Wei (1999). Only little is known about the per-
formance of the first for dependent test statistics [Benjamini, Hochberg and
Kling (1997)], and nothing about the second.
Finally, recall the resampling based procedure of Yekutieli and Benjamini
(1999), which tries to cope with the above problem and at the same time uti-
lize the information about the dependency structure derived from the sample.
The resampling based procedure is more powerful, at the expense of greater
complexity and only approximate FDR control.

APPENDIX

PROOF OF LEMMA 3.1. For each i E Io and increasing set D, we have to

show that

Pr(XeD IXi=x)
is increasing in x. We will achieve this by expressing

(31) Pr(X E D I Xi = X) = EU1xx=X Pr(X E D I Xi = x, U)

and showing that for x < x',

(32) Eulxi=x Pr(X E D I Xi = x, U) < Eulx=x, Pr(X E D I Xi = x', U).

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1184 Y. BENJAMINI AND D. YEKUTIELI
We prove the lemma in two steps.

1. For each x < x' we construct a new random variable U' whose marginal
distribution is stochastically smaller than the marginal distribution of U,
but its conditional distribution given Xi = x' is identical to the conditional
distribution of U given Xi = x.
2. We show that the newly defined random variable U' satisfies

(33) Pr(X E D I Xi = x, U = u) < Pr(X E D Xi = x', U' = u).

By re-expressing the second term in inequality (32) in terms of U' and then
using inequality (33), the proof is complete:

EulX=x, Pr(X E D I Xi = x', U) = EujX,=x,, Pr(X E D I Xi = x', U')

> Eux=x Pr(X ED I Xi = x, U).

STEP 1. The construction of U': according to condition (d) of this lemma,

U is PRDS on Xi; this means that the cdf of U I Xi = x' is less or equal to
the cdf of U I Xi = x,

(34) ~~~~Fulx,=x, < Fulx,=x.

In order to avoid technicalities let us assume that U I Xi = x has the same
support as U for any x. Now the following increasing transformation is well
defined, and satisfies

(35) hx,x (u) = F1x =x(Fuli=x,(u)) < F-x1x(Fuixx(u)) = u,

because of (34). The new random variable U' is defined as

U' = hx x/(U)

and is, from (35), stochastically smaller than U. Because g, Y and U are
continuous, the conditional distribution of U given Xi is continuous, hence
hx x and its inverse hx! x can be defined. Using the notation

(36) u' = hX, (U)

we can state the following properties:

(i) u < u', again because of (35), and hx, X being its inverse.
(ii) Fulx =x(u) = Fujx=x,(u'), which follows directly from the definition of
hx, xI
(iii) The events U < u' and U' < u are identical, as U' is a monotone
function of U.

Combining (i), (ii) and (iii), we get

Pr(U < u I Xi = x) = Pr(U < u' X = x')

= Pr(U' < u Xi = x').
Hence U I Xi = x and U' I Xi = x' are identically distributed.

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1185

STEP 2. A proof of inequality (33): the function gi is one-to-one, so the

values of U and Xi uniquely determine the value of Yi. Thus for each u, and
the corresponding u' defined in expression (36), denote y and y' those values
of Yi which satisfy

gi(y, u) = x and gi(y', u') = x'.

We now establish that for the pair x' > x, and the pair u' > u as above, we
also have that y' > y. As gi is strictly increasing in both components, fixing
Xi then Yi < y iff U > u, thus

Pr(Yi < y I Xi = x) = Pr(U > u Xi = x) = 1 - Fujx.=X(U)

Similarly, Yi < y' iff U > u',

Pr(Yi < y' I Xi = x') = Pr(U > u' Xi = x') = 1 - Fux,=x(u')

As Fujx1=x,(u') = Fujx =,(u), y and y' are quantiles corresponding to the
same probability. Returning to condition (d) of the lemma, Yi is PRDS on Xi,
therefore Yi I Xi = x' is stochastically greater than Yi I Xi = x, thus y < y'.
We now define

Y(D, u) := {Y: g(Y, u) E D}.

Note that if D is an increasing set then Y(D, u) is an increasing set. We can
now proceed to complete the proof of Step 2:

Pr(X E D I Xi = x, U = u) = Pr(Y E Y(D, u) Yi = y, U = u)

(37) < Pr(Y E Y(D, u) Yi = y', U =u)
(38) < Pr(Y E Y(D, u') Yi = y', U = u')
= Pr(X E D I Xi = x', U = u')

(39) =Pr(X E D Xi = x', U'= u)

Inequality (37) holds because Y is PRDS and independent of U. Using again
the independence, and the fact that if u < u' then Y(D, u) c Y(D, u'), we get
inequality (38). Finally as U' = u iff U = u' we get the equality in expression
(39). This completes the proof of Step 2, and thereby the proof of Lemma 3.2. 0

REMARK A. 1. Note that the seemingly simple route of proving Lemma 3.1
via showing

(40) Pr(X E D I Xi = x, U = u) < Pr(X E D I Xi = x', U = u)

does not yield the desired result, because the distribution of U I Xi = x is
different than the the distribution of U I Xi = x'.

REMARK A.2. In the course of the proof we established the monotonicity of

Pr(X eD Yi = y, U = u)

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1186 Y BENJAMINI AND D. YEKUTIELI
in y and in u. However, because gi is increasing, fixing Xi and increasing U
will decrease Yi, because Y is PRDS, and

(41) Pr(X E D I Xi = x, U = u)
does not necessarily increase in u. If expression 41 increases in u, for example
when the components of Y are independent, proof of Lemma 3.2 is immediate
because the distribution of U I Xi = x' is stochastically greater than the
distribution of U I Xi = x.

REMARK A.3. The assumption that U I Xi = x has the same support as

U is not critical. With appropriate definition of the inverse of the conditional
cdf of U, F1UIX hX X can be well defined over the entire range of U. Also hx, x
can be defined similarly. It will be the inverse of hx, x only on the respective
ranges. Properties (i)-(iii) still hold under this more complicated construction.

REMARK A.4. If conditions (a)-(c) of the lemma are met, while condition
(d), U and Yi, are PRDS on Xi is only true for Xi such that Xi > xi then
altering the proof accordingly, X is PRDS on Xi > xi.

Acknowledgments. We are grateful to Ester Samuel-Cahn, Yosef Rinott

and David Gilat for their helpful comments and to a referee for keeping us
honest.

REFERENCES

ABRAMOVICH, F. and BENJAMINI, Y. (1996). Adaptive thresholding of wavelet coefficients. Comput.

Statist. Data Anal. 22 351-361.
BARINAGA, M. (1994). From fruit flies, rats, mice: evidence of genetic influence. Science 264
1690-1693.
BENJAMINI, Y. and HOCHBERG, Y. (1995). Controlling the false discovery rate: a practical and
powerful approach to multiple testing. J Roy. Statist. Soc. Ser. B 57 289-300.
BENJAMINI, Y. and HOCHBERG, Y. (1997). Multiple hypotheses testing with weights. Scand. J
Statist. 24 407-418.
BENJAMINI, Y. and HOCHBERG, Y. (2000). The adaptive control of the false discovery rate in mul-
tiple hypotheses testing. J Behav. Educ. Statist. 25 60-83.
BENJAMINI, Y., HOCHBERG, Y. and KLING, Y. (1993). False discovery rate control in pairwise com-
parisons. Working Paper 93-2, Dept. Statistics and O.R., Tel Aviv Univ.
BENJAMINI, Y., HOCHBERG, Y. and KLING, Y. (1997). False discovery rate control in multiple
hypotheses testing using dependent test statistics. Research Paper 97-1, Dept. Statis-
tics and O.R., Tel Aviv Univ.
BENJAMINI, Y. and WEI, L. (1999). A step-down multiple hypotheses testing procedure that
controls the false discovery rate under independence. J Statist. Plann. Inference 82
163-170.
CHANG, C. K., ROM, D. M. and SARKAR, S. K. (1996). A modified Bonferroni procedure for repeated
significance testing. Technical Report 96-01, Temple Univ.
EATON, M. L. (1986). Lectures on topics in probability inequalities. CWI Thact 35.
HOCHBERG, Y. (1988). A sharper Bonferroni procedure for multiple tests of significance.
Biometrika 75 800-803.

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
 CONTROLLING THE FDR UNDER DEPENDENCY 1187

HOCHBERG, Y. and HOMMEL, G. (1998). Step-up multiple testing procedures. Encyclopedia Statist.
Sci. (Supp.) 2.
HOCHBERG, Y. and ROM, D. (1995). Extensions of multiple testing procedures based on Simes'
test. J Statist. Plann. Inference 48 141-152.
HOCHBERG, Y. and TAMHANE, A. (1987). Multiple Comparison Procedures. Wiley, New York.
HOLLAND, P. W. and ROSENBAUM, P. R. (1986). Conditional association and unidimensionality in
monotone latent variable models. Ann. Statist. 14 1523-1543.
HOLM, S. (1979). A simple sequentially rejective multiple test procedure. Scand. J. Statist 6 65-70.
HOMMEL, G. (1988). A stage-wise rejective multiple test procedure based on a modified Bonferroni
test. Biometrika 75 383-386.
Hsu, J. (1996). Multiple Comparisons Procedures. Chapman and Hall, London.
KARLIN, S. and RINOTT, Y. (1980). Classes of orderings of measures and related correlation
inequalities I. Multivariate totally positive distributions. J. Multivariate Statist. 10
467-498.
KARLIN, S. and RINOTT, Y. (1981). Total positivity properties of absolute value multinormal
variable with applications to confidence interval estimates and related probabilistic
inequalities. Ann. Statist. 9 1035-1049.
LANDER E. S. and BOTSTEIN D. (1989). Mapping Mendelian factors underlying quantitative traits
using RFLP linkage maps. Genetics 121 185-190.
LANDER, E. S. and KRUGLYAK L. (1995). Genetic dissection of complex traits: guidelines for inter-
preting and reporting linkage results. Nature Genetics 11 241-247.
LEHMANN, E. L. (1966). Some concepts of dependence. Ann. Math. Statist. 37 1137-1153.
NEEDLEMAN, H., GUNNOE, C., LEVITON, A., REED, R., PRESIE, H., MAHER, C. and BARRET, P. (1979).
Deficits in psychologic and classroom performance of children with elevated dentine
lead levels. New England J Medicine 300 689-695.
PATERSON, A. H. G., POWLES, T. J., KANIS, J. A., MCCLOSKEY, E., HANSON, J. and ASHLEY, S. (1993).
Double-blind controlled trial of oral clodronate in patients with bone metastases from
breast cancer. J Clinical Oncology 1 59-65.
ROSENBAUM, P. R. (1984). Testing the conditional independence and monotonicity assumptions of
item response theory. Psychometrika 49 425-436.
SARKAR, T. K. (1969). Some lower bounds of reliability. Technical Report, 124, Dept. Operation
Research and Statistics, Stanford Univ.
SARKAR, S. K. (1998). Some probability inequalities for ordered MTP2 random variables: a proof
of Simes' conjecture. Ann. Statist. 26 494-504.
SARKAR, S. K. and CHANG, C. K. (1997). The Simes method for multiple hypotheses testing with
positively dependent test statistics. J. Amer. Statist. Assoc. 92 1601-1608.
SEEGER, (1968). A note on a method for the analysis of significances en mass. Technometrics 10
586-593.
SEN, P. K. (1999a). Some remarks on Simes-type multiple tests of significance. J Statist. Plann.
Inference, 82 139-145.
SEN, P. K. (1999b). Multiple comparisons in interim analysis. J Statist. Plann. Inference 82
5-23.
SHAFFER, J. P. (1995). Multiple hypotheses-testing. Ann. Rev. Psychol. 46 561-584.
SIMES, R. J. (1986). An improved Bonferroni procedure for multiple tests of significance.
Biometrika 73 751-754.
STEEL, R. G. D. and TORRIE, J. H. (1980). Principles and Procedures of Statistics: A Biometrical
Approach, 2nd ed. McGraw-Hill, New York.
TAMHANE, A. C. (1996). Multiple comparisons. In Handbook of Statistics (S. Ghosh and C. R. Rao,
eds.) 13 587-629. North-Holland, Amsterdam.
TAMHANE, A. C. and DUNNETT, C. W. (1999). Stepwise multiple test procedures with biometric
applications. J Statist. Plann. Inference 82 55-68.
TROENDLE, J. (2000). Stepwise normal theory tests procedures controlling the false discovery rate.
J. Statist. Plann. Inference 84 139-158.
WASSMER, G., REITMER, P., KIESER, M. and LEHMACHER, W. (1999). Procedures for testing multiple
endpoints in clinical trials: an overview. J. Statist. Plann. Inference 82 69-81.

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms
1188 Y BENJAMINI AND D. YEKUTIELI
WELLER, J. I., SONG, J. Z., HEYEN, D. W, LEWIN, H. A. and RON, M. (1998). A new approach to the
problem of multiple comparison in the genetic dissection of complex traits. Genetics
150 1699-1706.
WESTFALL, P. H. and YOUNG, S. S. (1993). Resampling Based Multiple Testing, Wiley, New York.
WILLIAMS, V. S. L., JONES, L. V. and TUKEY, J. W (1999). Controlling error in multiple comparisons,
with special attention to the National Assessment of Educational Progress. J Behav.
Educ. Statist. 24 42-69.
YEKUTIELI, D. and BENJAMINI, Y. (1999). A resampling based false discovery rate controlling mul-
tiple test procedure. J Statist. Plann. Inference 82 171-196.

SCHOOL OF MATHEMATICAL SCIENCES

DEPARTMENT OF STATISTICS
AND OPERATIONS RESEARCH
TEL AvIv UNIVERSITY
RAMAT AvIv, 69978 TEL AvIv
ISRAEL
E-MAIL: [email protected]
[email protected]

This content downloaded from 61.231.157.215 on Fri, 30 May 2025 09:45:48 UTC
All use subject to https://about.jstor.org/terms