Ladder Heights, Gaussian Random Walks, and The Riemann Zeta Function

Ladder Heights, Gaussian Random Walks,
and the Riemann Zeta Function

y
By Joseph T. Chang and Yuval Peres
Yale University and University of California at Berkeley

July 23, 1996
Let fSn : n 0g be a random walk having normally distributed increments with

mean and variance 1, and let be the time at which the random walk rst takes a
positive value, so that S is the rst ladder height. Then the expected value E S ,
originally dened for positive , may be extended to be an analytic function of the
complex variable throughout the entire complex plane, with the exception of
certain branch point singularities. In particular, the coecients in a Taylor
expansion about = 0 may be written explicitly as simple expressions involving the
Riemann zeta function. Previously only the rst coecient of the series developed
here was known; this term has been used extensively in developing approximations
for boundary crossing problems for Gaussian random walks. Knowledge of the
complete series makes more rened results possible; we apply it to derive
asymptotics for boundary crossing probabilities and the limiting expected overshoot.
AMS 1991 subject classications. 60J15, 30B40.

Key words and phrases. Random walk, ladder height, Riemann zeta function, boundary crossing
probability, analytic continuation.
y Research partially supported by NSF grant # DMS-9404391.
1
Ladder Heights and the Zeta Function 2
1 Introduction
Let X ; X ; : : : be independent and distributed as N (; 1), the normal distribution
1 2
with mean and variance 1. Consider the random walk fSn : n 0g consisting of the
partial sums Sn = X + + Xn. The rst time = inf fn : Sn > 0g that the random
1
walk is positive is called the rst ladder epoch, and the rst positive value S taken by the
random walk is called the rst ladder height.
The Riemann zeta function is originally dened for Re(z) > 1 by the series
(z ) = 1 ?z
n=1 n . This denition is extended by analytic continuation to the entire
P
complex plane except z = 1, where has a simple pole. For example, the extension to
Re(z) < 0 may be seen from Riemann's functional equation (see (19)), which expresses
(z ) in terms of (1 ? z ). Calculation of the zeta function is routine; it is implemented in
the computer package Mathematica, for example.
For positive , the expected value of S depends on ; let us write it as E S to

indicate this dependence. The main subject of this paper is the behavior of E S near
= 0. In particular, we show that the function 7! E S can be continued analytically
to a neighborhood of the origin, and we nd its Taylor expansion about = 0. The next
theorem is our main result.
Theorem 1.1 The expected rst ladder height E S may be extended to be an

analytic function in a neighborhood of = 0; it has the expansion
1 ? X
(
1 ( ? n) ?
1 2 n
! )
E S = p exp p 2
(1)
2 2 n n!(2n + 1) 2
=0
p
in the disk fjj < 2 g.
Previously only the n = 0 term in the expansion of Theorem 1.1 was known explicitly.
We will also determine the extent to which the function 7! E S can be analytically
continued throughout the complex plane. It turns out that the function has branch points
p p
at f 2k i 2k : k 6= 0; k integerg; in particular, it can be analytically continued to a
neighborhood containing the real axis.
Our starting point is a classical expression for E S from the uctuation theory of
random walks; see (5) below. As # 0, the series in (5) converges ever more slowly,
whereas the convergence in the Taylor expansion (1) that we shall derive from it becomes
more rapid. The behavior of E S as # 0 is of particular signicance for applications
such as heavy trac limit theorems for queues [see, for example, Whitt (1974) and
Asmussen (1987)] and the corrected diusion approximations of Siegmund (1979).
Theorem 1.1 is based on a result (see Theorem 2.1 in Section 2) about power series of
the form P1 ? ?nw , which also arise in other contexts. For each this series
n n e =1
represents an analytic function of w in the right half-plane fRe(w) > 0g. We investigate
the analytic continuation of this function; we are particularly interested in continuing to a
neighborhood of w = 0. Most of the work was in fact done by Hardy (1905), who
investigated certain questions about power series, including the behavior of the series
P 1 n? z n about the point z = 1. So our problem involves a reparametrization of
n=1
Hardy's. This reparametrization turns out to be felicitous, in that the resulting expansion
has a neat and explicit expression in terms of the zeta function: whereas Hardy stopped
with the rst coecient in his expansion, we obtain explicit expressions for all of the
coecients in ours. To keep the present paper self contained, some of our development
will closely follow Hardy's.
For simplicity, our treatment here is conned to random walks with Gaussian
increments, which is the case that has been the most important in applications. The
Gaussian case also plays a central role in the theory; a heuristic sense of this may be
obtained by combining the central limit theorem with the basic identity (5) below, which
holds for general increment distributions. An intricate analysis in this spirit was carried
through by Lai (1976), who analyzed more general increment distributions|although
with the additional restriction of zero mean|by expanding around the Gaussian case.
Likewise, the importance of the series P1 ? n
n n z goes beyond the Gaussian case; indeed
=1
a two-term asymptotic expansion of this series as z increases to 1 forms a cornerstone of

the analysis of Lai (1976).
Moments of the rst ladder height have been studied in the general setting of an
exponential family of distributions fF : 2 g, which may be written in the form
F (dx) = ex?
( )
F0 (dx). For example, the normal family fN (; 1) : 2 IRg under
consideration here is a special case having () = =2. Let P and E denote probability
2
and expectation when the distribution of Xi is F . Under the assumptions that

contains an interval around 0, E X = 0, and the distribution F is nonlattice, Siegmund
0 1 0
(1979) showed that for a > 0,

a
E Sa = E0 Sa + (E Sa ) + o()
+1
(2)
a+1 0
as # 0. The question of whether the remainder o() in (2) is actually O( ) was left
2
open by Siegmund's development. Under the slightly strengthened hypothesis that the
distribution F is strongly nonlattice, Chang (1992) replaced the o() by ca + O( ).
0
2 3
Here ca is a constant dened in terms of F whose general form is rather complicated, but
0
it takes a simple form when a = 1 and the distribution F is symmetric about 0. For the
0
normal family, these previous results gave
p1 exp + O( ) ;
h i
E S = 3
(3)
2
where = E (S )=(2E S ) =: 0:583.
0
2
0
The quantity has an interesting history. It arose in a paper of Cherno (1965), who
studied discrete and continuous versions of the problem of sequentially testing whether
the drift of a Brownian motion is positive. In the discrete version corresponding to a
given > 0, the Brownian motion is observed only at times 0; ; 2; : : :. Both the discrete
and continuous versions have optimal solutions of the form: stop the Brownian motion at
the rst observation that crosses a certain boundary. Denoting the optimal boundaries for
the discrete problem and the continuous problem by x = x (t) and x = x(t), respectively,
Cherno (1965) showed, in his notation, that
p p
x (t) = x(t) + z^ + o( ) as # 0: (4)
It turns out that the number z^ is in fact ?. A more fundamental probabilistic
interpretation of is as a limiting expected overshoot: dening the overshoot
Rb = S (b) ? b, where (b) is the rst passage time inf fn : Sn > bg, standard results from
renewal theory say that Rb converges in distribution to a random variable R1, and
E0 R1 = E0 (S2 )=(2E0 S ). Siegmund's result (2) with a = 1 shows that the limiting
expected overshoot is also the coecient of in the expansion of E S =E S as # 0.
0
As for numerical computation, Cherno (1965) and Siegmund (1979) expressed as an

integral and Lai (1976) expressed as the sum of a series. At the time their work was
done, Lai and Siegmund were not aware of the connection to the work of Cherno; Hogan
(1986) explained the connection between the solution of Cherno's problem and the
limiting expected overshoot.
As far as we are aware, the only previous observation that relates Gaussian random
walks and the zeta function is the tantalizing nal sentence of the following quotation
from Cherno (1965) about his result (4):
This result involves relating the original problem to an associated problem

and studying the limiting behavior of the solution of the associated problem.
This solution corresponds to the solution of a Wiener-Hopf equation. Results
of Spitzer can be used to characterize the solution of the Wiener-Hopf
equation and yield z^ as an integral, which, as Gordon Latta pointed out to the
author, is equal to (1=2)=(2) = = ?:5824.
1 2
[References omitted. Also, the quoted numerical value is slightly incorrect; to 4 decimal
places it should be ?:5826.] That = ? (1=2)=(2) = can be seen by comparing (3) and
1 2
the expansion in Theorem 1.1. Lai (1976) partially rediscovered the expansion of Hardy
by a dierent method. However, Lai's development recovered only the same terms that
Hardy gave explicitly, and did not establish analyticity or relate the results to the zeta
function.
The rst ladder height plays a fundamental role in the theory and applications of
random walks and renewal theory; see, for example, the treatises of Asmussen (1987),
Feller (1971), Prabhu (1980), and Siegmund (1985), all of which feature ladder variables
prominently. The particular area of application that originally motivated this
investigation is the asymptotic approximation of boundary crossing probabilities for

random walks. In addition to certain areas of applied probability such as queueing theory
(e.g. Asmussen, 1987) and insurance risk analysis (Grandell, 1991), such boundary
crossing problems arise in statistics from the study of sequential procedures, which
typically sample until a certain random walk crosses a certain boundary; see Siegmund
(1985) and Woodroofe (1982). They also arise in some nonsequential procedures such as
likelihood ratio tests for a change point (e.g. Siegmund, 1986); here the likelihood ratio
involves a random walk indexed by the possible change points, and the test rejects if the
maximum of the likelihood ratio exceeds some level, so that signicance levels and
P -values are boundary crossing probabilities. Some of the most useful and accurate
approximations are derived by taking a limit as the drift of the random walk tends to
zero and the boundary tends to innity. Knowledge of the expected amount by which a
random walk overshoots a high level is a basic requirement for the development of this
theory. Since this limiting expected overshoot is E (S )=(2E S ), to obtain asymptotics
2
for the overshoot as ! 0, the behavior of moments of the rst ladder height as ! 0
must be understood.
The analytic continuation and Taylor expansion of log E S are developed in the
next section. In the nal section we discuss some simple applications to the study of
boundary crossing probabilities.
2 Analytic Continuation of ES

The relation
1 1
P fSn 0g for > 0
X
log E =
n=1 n
is a standard fact in the uctuation theory of random walks; see, e.g., Theorem 3 on p.
416 of Feller (1971). Wald's equation E S = E gives
1 1
!
P fSn 0g :
X
E S = exp (5)
n=1 n
Dierentiating (5) with respect to , which can be justied by dominated convergence,

and using normality of Sn, we obtain
d 1 1 1 (e? =2 )n 2
log E S = ? p pn
X
(6)
d 2 n=1
for > 0. This motivates our study of series of the form P n?e?nw ; we are particularly
interested in the case = 1=2 and w = =2, and we would like to determine the behavior
2
of the series around w = 0. This is done in Theorem 2.1 below.
Notation: It will be convenient to dene c() = 2i=?(1 ? ). Since the Gamma

function ?(z) is analytic at all z 2 C except the points z = 0; ?1; ?2; : : :, where it has
simple poles, it follows that c() is an entire function, with zeros at positive integer values
of . From the relation ?()?(1 ? ) = = sin() we obtain the alternative expression
c() = 2i?() sin() for all complex other than the nonpositive integers 0; ?1; ?2; : : :.
Theorem 2.1 For each 2 C, the function

1
n? e?nw ? 2iw?1 ;
X
H (w) := c() (7)
n=1
which is analytic in the half-plane fRe(w) > 0g, may be analytically continued to the disk
fjwj < 2g. In that disk, for each 2 C other than the positive integers, H has the
Taylor series expansion
1 (?w)n
( ? n)
X
H (w) = c() : (8)
n=0 n!
Remark: With the appropriate interpretation, (8) holds also when is a positive
integer. In this case, the product c() ( ? n) should be interpreted as 0, except for the
case n = ? 1, where one uses that fact that c() ( ? n) has a removable singularity at
= n + 1.
Theorem 1.1 is an easy consequence: taking = 1=2 and w = =2 and using the fact 2
p p
that ?(1=2) = (so that c(1=2) = 2i ) gives
1 (e?2 =2 )n p ?=
1 2 1 (1=2 ? n) ? 2 n
!
pn ? 2
X X
2
= :
n=1 n=0 n! 2
Combining this with (6) we get
log E S = p?1
d X1 (1=2 ? n) ? 2 n
!
:
d 2 n=0 n! 2
Theorem 1.1 is obtained by integrating, exponentiating, and using the fact that
p
E0 S = 1= 2.
Proof of Theorem 2.1. Removing the ray IR? = fx 2 IR : x 0g from the

complex plane leaves a domain C n IR? on which the logarithm function may be dened
to be single-valued and analytic by
log u = log juj + iarg(u);

Figure 1.
Figure 2.
where ? < arg(u) < , say. We then adopt the denition uz = ez log u; for each z 2 C this
is also an analytic function of u for u 2 C n IR?. Having xed a branch of the logarithm,
certain operations require a bit of care; for example, a fact that we will use below is
(xy)z = xz yz if jarg(x) + arg(y)j < : (9)
For small positive , let C denote the contour
C = fx + i : ? ?1 x 0g + fei : =2 ?=2g + fx ? i : 0 x ? ?1 g;
traversed as shown in the dashed curve in Figure 1. Here the addition of contours
indicates successive traversal in the positive direction (i.e. whatever direction the contour
had as originally dened). Taking a > 0 and b 2 (0; 2), let
L = f?? + iy : ? y b ? 2g + fx + i(b ? 2) : ?? x ag

1 1
+ fa + yi : b ? 2 y bg + fx + bi : a x ? ? g + f?? + iy : b y g;
1 1
this is the solid contour in Figure 1. Further, let
L = fx + i(b ? 2 ) : ?1 < x ag + fa + yi : b ? 2 y bg + fx + bi : a x > ?1g
and let L0 denote the dierence L ? L , as shown in Figure 2. Finally, let D denote the
domain
D = fu : Re(u) < a and b ? 2 < Im(u) < bg
bounded by L, and let D be a bounded domain containing the point 0 such that the
0
closure of D is contained in D.
0
Lemma 2.2 (Hardy) Let H be as dened in (7), and dene

Z
u?1
I (w) := e?w du: (10)
L e?u ? e?w
Then for each a 2 C, H (w) is an analytic function of w for Re(w) > 0, and I (w) is
analytic for w 2 D. For all w 2 D \ fRe(w) > 0g and Re() > 0 we have H (w) = I (w).
Proof. Let be arbitrary. Clearly H (w) is an analytic function of w for

Re(w) > 0, since the series in the denition (7) converges uniformly in fRe(w) cg for all
c > 0. Dening = supfRe(?w) : w 2 D0 g, for each 2 (0; 1=) the integral
Z
u?1
du (11)
L e?u ? e?w
is easily seen to be an analytic function of w for w 2 D ; for example, one could use a
0
combination of Fubini's and Morera's theorems, or section 2.83 of Titchmarsh (1939).

Furthermore, for w 2 D ,0
u?1 ju?1j

e?u ? e?w je?u j ? e

holds for Re(?u) > . Thus,

Z
u?1
du ! 0 as ! 0;
L0 e?u ? e?w
with the convergence being uniform over w 2 D . Therefore, 0
Z
u?1 Z
u?1
lim du = du (12)
!0 L e?u ? e?w L e?u ? e?w
is analytic in w on the domain D , so that, by the arbitrary nature of D , (12) is analytic

0 0
on all of D, and hence so is I.
Next let Re() > 0. A change of variables in the denition ?() = 1 x? e?x dx

R
1
0
gives
n? =
1 Z juj? enu du
0
1
?() ?1
for n 1. From this, for Re(w) > 0 we obtain
1 e?w Z 0 juj?1
n? e?nw
X
= du; (13)
n=1 ?() ?1 e?u ? e?w
by Fubini's theorem.
For w inside the closed contour C + L ,

Z
u?1
du = ?2iew w?1 : (14)
C +L e?u ? e?w
This can be seen by integrating over a small circle about w, or by changing variables and
using Cauchy's integral formula. Therefore, for each w 2 D n IR? , (14) holds for
suciently small positive .
Next observe that for Re() > 0,
lim
Z
u?1
du = ?2i sin( )
Z 0
juj?1 du for w 2 D n IR : (15)
!0 C e?u ? e?w
?
?1 e?u ? e?w
Indeed, since (u i)? ! juj? exp[i( ? 1)] for u < 0 as # 0, routine dominated
1 1
convergence arguments show that for w 2 D n IR?,
lim
Z 0
(u i)? du = ?ei Z
1 0 juj? du:1
#0 ?(1=) e ? ui

( w)
?e ?1 e?u ? e?w
Thus, (15) follows from the statement

Z ?=2 (ei )? 1
iei d ! 0 as # 0;
=2 e?ei ?e ?w
which holds for Re() > 0.
Therefore, by (12), (14), and (15),

Z
u?1
?2iew w? = lim
1
! 0 C +L e?u ? e?w
du
= ?2i sin()
Z 0
juj?1
du +
Z
u?1
du
?1 e?u ? e?w L e?u ? e?w
for w 2 D n IR? and for Re() > 0. Combining this with (13) gives H(w) = I(w) for
w 2 D \ fRe(w) > 0g provided that Re() > 0.
A key idea, also used by Hardy, is to consider the analyticity of various expressions
as functions of the variable as well as w. In the proof of the theorem, we will establish
certain identities rst for values of with large real part, then argue by analytic
continuation that the identities hold for other values of . The next lemma prepares the
way.
Lemma 2.3 Let H and I be as dened in (7) and (10). Then for each
w 2 D \ fRe(w) > 0g, the functions 7! H (w) and 7! I (w) are both entire. Also, for
each nonnegative integer k, the kth derivative

dk
I(k) (0) = k I (w)
dw w=0
is an entire function of .
Proof. For Re(w) > 0, the sum 1 ? e?nw converges uniformly over in
P
1 n
bounded sets, so the sum is an entire function of , and hence so is 7! H (w). For
w 2 D, the integral (11) is an analytic function of for small enough positive , and the
convergence in (12) is uniform over in bounded sets. Therefore, 7! I(w) is entire in
.
To prove the nal assertion of the lemma we will use the expansion
e?w 1 +X 1 e?u
( )
= (e ?w ? 1)n for je?w ? 1j < je?u ? 1j:

e?u ? e?w e?u ? 1 n=1 (e?u ? 1)n+1
Letting be arbitrary, note that if we take w in a small enough neighborhood N of 0 so
that j(e?w ? 1)=(e?u ? 1)j r < 1 holds for all w 2 N and u 2 L, then
1 (e?w ? 1)n ju? j 1 e?w ? 1 n jduj

Z Z 1
X X
L
ju? j
1

(e?u ? 1)n
+1
jduj =
L je?u ? 1j n

e?u ? 1

=0 n =0

1
X
rn
Z
ju? j jduj < 1:
1
n =0 L je?u ? 1j
Therefore, we may interchange sum and integral to obtain

Z
u?1 Z
u?1 1 Z
u?1 e?u
I (w) = e?w du = du +
X
(e ?w ? 1)n du: (16)
L e?u ? e?w L (e?u ? 1) n=1 L (e?u ? 1)n+1
for w 2 N . The convergence is uniform over w 2 N .
To evaluate the derivatives Ik (0), write

( )
!
dk ?w n n
? (?j )k (?1)n?j ;
X
(e 1) n =
dwk
w=0 j =0 j
this is 0 if n > k. Since (16) expresses I (w) as a uniformly convergent series of analytic
functions in the neighborhood N , we may dierentiate term by term to obtain
u?1 e?u
( ) !
dk k Z n n
(?j )k (?1)n?j :
X X
I (w )
= du (17)
dwk w=0 n=1 (e?u ? 1)n

L +1
j =0 j
However, for each n > 0, the integral
Z
u?1 e?u
du (18)
L (e?u ? 1)n +1
denes an entire function of , again by an easy combination of Fubini's and Morera's

theorems. So the desired derivatives in (17) are entire in .
To complete the proof of Theorem 2.1, by Lemma 2.2, for each 2 C we have
X1 wk
I (w) = I(k) (0)
k=0 k!
for all w in a neighborhood of 0. The idea will be to evaluate the derivatives Ik (0) by ( )
using the function H .
Let w 2 D \ fRe(w) > 0g. Then Lemma 2.2 gives H(w) = I(w) for all with
Re() > 0. Therefore, by Lemma 2.3 and analytic continuation, H (w) = I (w) holds for
all a 2 C.
Next let be arbitrary. Since we have just shown that H(w) = I(w) for all
w 2 D \ fRe(w) > 0g, and since both H () and I () are analytic on D \ fRe(w) > 0g,
for each nonnegative integer k we have the equality of the kth derivatives
H(k) (w) = I(k) (w) for w 2 D \ fRe(w) > 0g:
However, by Lemma 2.2, each derivative Ik (w) is an analytic function of w for w 2 D.

( )
Therefore, we may compute Ik (0) as the limit

( )
I(k) (0) = lim I (k) (w) = lim

w#0
H (k) (w)
w#0
as w decreases to 0 along the real axis. Performing the last dierentiation gives
(
1 )
(? 1)k c() n?(?k) e?nw ? 2i( ? 1) ( ? k)w?k?1

X
Ik
( )
(0) = lim
w#
0
;
n=1
which, for Re() > k + 1, is obviously (?1)k c() ( ? k). Finally, by Lemma 2.3 and
analytic continuation, the equality Ik (0) = (?1)k c() ( ? k) holds for throughout
( )
the entire complex plane, except for the (removable) singularity of c() ( ? k) at

= k + 1. Thus, for other than positive integers, I , which is the analytic continuation
of H to a neighborhood of 0, has the Taylor series claimed in the theorem.
By the functional equation of Riemann (see, e.g., section 4.44 of Titchmarsh, 1939),
which expresses (1 ? z) in terms of (z) for z 6= 1,
( ? k)
= 2?k ?k?1 cos[(? + k + 1)=2] ?(k + 1 ? ) (k + 1 ? ): (19)
k! k!
Since ?(k + 1 ? )=(k!) k? and (k + 1 ? ) ! 1 as k ! 1, it is easy to see that the
radius of convergence of the power series P ( ? k)(?w)k =(k!) is 2. This proves
Theorem 2.1.
Next we investigate the complete analytic continuation of the function 7! E S ; we

will see that it may be continued to the whole plane, with countably many rays removed.
Theorem 2.4 For each integer J 0, the dierence

1 J
n? e?nw ? 2i (w ? 2ji)? ;
X X
c() 1
n=1 j =?J
which is analytic in the half-plane fRe(w) > 0g, may be analytically continued to the disk
fjwj < 2(J + 1)g, where it has the Taylor expansion
1 ?1
!
w k
sin[( ? k)=2]J;(k)
X
2i(2) ; (20)
k=0 k 2
where J; (k) = (k + 1 ? ) ? J j ?(k+1?) .
P
j =1
Proof. Start by writing
?(k + 1 ? ) = (?1)k ( ? 1)( ? 2) ( ? k)?(1 ? );
so that
?(k + 1 ? ) = (?1)k ? 1 ?(1 ? ) = (?1)k ? 1 2i :
! !
k! k k c()
Further substituting [ei k ? = + e?i k ? = ] for cos[(? + k + 1)=2] in (19), we
1
2
( +1 ) 2 ( +1 ) 2
obtain
1 ( ? k)
(?w)k = (2) fi T? ? (?i) T g
X
c() (21)
k=0 k! +
where
1 ?1 iw k (k + 1 ? ):
!
X
T =
k=0 k 2
For J 0, use the denition of J; (k) to write

1 ?1 iw k (k) + 1 ?1 iw k ;
! ! !
J
j ?1
X X X
T = J;
k=0 k 2 j =1 k=0 k 2j
this involves an interchange of order of summation that is easily justied if jwj < 2. By
the binomial theorem,
?1
1 ?1 iw k
! ! !
J J
iw
(2)ei=2 j ?1 = (2i) j ?1 1
X X X
j =1 k=0 k 2j j =1 2j

J
= 2i (w 2ji)? :
X
1
j =1
In the last equality we have used (9) together with the assumption that jwj < 2. Thus,
for jwj < 2,
1 ( ? k) J J
(?w)k ? 2i (w + 2ji)? ? 2i (w ? 2ji)?
X X X
c() 1 1
k=0 k! j =1 j =1
i=2 X ? 1 ?iw k
(
1 !
1 ? 1! iw k )
(k) ? e ? X
= (2) e J; i=2
J; (k )
k=0 k 2 k=0 k 2
1 ?1
!

w k
J; (k) sin[( ? k)=2];
X
= (2)2i
k=0 2 k
so that, by Theorem 2.1,

1 J
n? e?nw ? 2i (w ? 2ji)?
X X
c() 1
k=0 j =?J
1 ?1
!
X w
k
= ?2i(2) J; (k) sin[(k ? )=2]

k=0 2 k
holds for fjwj < 2; Re(w) > 0g. For xed J , simple estimates show that
J; (k) (J + 1)? k ? ( +1 Re( )) as k ! 1;
so that the last power series has radius of convergence 2(J + 1).
In summary, the function G(w) = c() P1 ? ?nw can be continued throughout

n n e =1
the complex plane except for branch point singularities at the points 2ki for integers k
(and so we must remove a ray, say Rk = fx + 2ki : x 0g, from the plane for each
branch point 2ki). Since G is periodic (with period 2i) in the right half-plane
fRe(w) > 0g, its analytic continuation is clearly periodic in the domain C n 1k ?1 Rk . S
=
The branch point singularities of G are of the form B;k (w) = 2i(w ? 2ki)? ; that is, 1
for each k, the dierence G ? B;k can be analytically continued to a neighborhood of

2ki. The expected rst ladder height involves the dierence G = ? B = ; in a very
1 2 1 20
simple way; in fact,

( ! !)
d
log E S = pi G1=2 2
2
2
?B = ; 2 :
d 2 2 1 20
However, we know that G = (w) ? B = ; (w) may be analytically continued to all w in the
1 2 1 20
domain C n Sk6 Rk . Thus, E S may be continued for throughout the complex plane,
=0
p
except for branch point singularities at the points = 2 kei= for nonzero integers k
4
(and the attendant rays removed). In particular, E S is clearly continuable to a

horizontal strip that contains the real axis.
Finally, we remark that in any xed disk fjwj < cg, as J increases, the series in (20)
converges more and more rapidly, so that Theorem 2.4 provides a computationally
ecient method of calculating G(w).
3 Applications
Perhaps the most fundamental of boundary crossing problems for random walks is
determining the probability that a random walk with negative drift ever attains a positive
level b. Letting (b) denote the rst passage time inf fn : Sn > bg, we want to
approximate the probability P? f (b) < 1g, where > 0.
Corollary 3.1 There exists r > 0 such that, for 0 < 2p,
(
1 1 ( ? n) ? 1 2 n
! )
P? f (b) < 1g = exp ?2 b ? p + O(e?rb)

X
2
2 n n!(2n + 1) 2
=0
as b ! 1.
Proof. The overshoot (or residual ) Rb over the level b is dened by Rb = S (b) ? b
when (b) is nite. By Wald's likelihood ratio identity (e.g. Siegmund (1985) page 13),
P? f (b) < 1g = E (e?2S b ) = e?2b E (e?2Rb ):

( )
(22)
Under P , we have (b) < 1 with probability 1, and, as b tends to innity, Rb converges
in distribution to a random variable R1 having probability density P fS > xg=(E S )
with respect to Lebesgue measure for x > 0. The expectation E (e? Rb ) converges to
2
() := E (e?2R1 ), where we have adopted the notation used by Siegmund. Corollary
2.3 of Chang (1992) shows that this convergence is exponentially fast: for > 0, there
exist C < 1 and r > 0 such that
jE (e? R ) ? ()j Ce?rb

2 b
(23)
for all 2 [0; ] and all b 0. The fact that > 0 may be chosen arbitrarily follows
from the fact that the family of distributions fN (; 1) : 0 g is uniformly strongly
nonlattice, that is,
0
inf inf j1 ? E (eitX )j > 0 for all > 0:
jtj
We have
1 1 ?2x
Z
1 1 [1 ? E (e?2S )]:
() = e P fS > xg dx =
E S 0 E S 2
However, letting ? denote the rst descending ladder epoch inf fn : Sn 0g, Wald's
likelihood ratio identity, symmetry of the normal distribution, and duality (e.g. Corollary
8.39 of Siegmund, 1985) give
1 ? E (e? S ) = 1 ? P? f < 1g = P f? = 1g = 1=(E );

2
so that
1 1
() = = :
(E S )(2E ) 2(E S )2
From this, our Taylor series for log E S may be transformed into a Taylor series for
log () about = 0, resulting in
8 s 9
2X 1 ( ? n) ?
< 1 2
! n=
() = exp : 2
: (24)
n n!(2n + 1) 2
=0
;
Combining this with (22) and (23) proves the corollary.
For example, in a diusion normalization, which assumes that ! 0 and b ! 1

with b 1, clearly the error term O(e?rb) is dominated by each power of . More
generally, we could let ! 0 and b ! 1 much slower than in a diusion normalization,
as in the following statement.
Corollary 3.2 If # 0 and b ! 1 in such a way that

b
log(1=)
! 1;
then
(
1 X 1 ( ? n) ?
1 2 n )
!
P? f (b) < 1g exp ?2 b ? p 2

:
2 n n!(2n + 1) 2
=0
The meaning of the last display is as an asymptotic expansion in the sense of Poincare:
that is, for each N ,
(
1 N ( ? n)

? 1 2 n
! )
P? f (b) < 1g = exp ?2 b ? p

X
2
+ O( N +2 )
2
:
2 n n!(2n + 1) 2 =0
Other results that follow easily from Theorem 2.1 are explicit expressions for the
moments E (Sp). These may be found by calculating along the lines shown by Lai (1976),
0
using our expansion of the series P n?e?nw about w = 0; here the series with values of
other than = 1=2 arise. For example, the results for the third and fourth moments are
!
E (S ) = p
3 3 1 + (1=2) = 1:25035
2
0
2 4 2
and
(3=2) 3(1=2) 3 (1=2)
E0 (S4 ) =
3=2
? 3=2 ? 2p = 2:26433:
For another example of information that can be extracted from Theorem 1.1, dene
p
the function () = E R1; note that the value of (0) is the number ? (1=2)= 2 we
have been calling . The rst derivative 0(0) = 1=4 is known. The second derivative
00 (0) can be obtained by equating the expansion (24) for the function () with its
denition E (e? R1 ), writing the Taylor expansion of the exponential up to order , and
2 2
using (2). The result of this calculation, which also makes use of the moments E (S ) and 0
3
E0 (S4 ) given in the previous paragraph, is

(3=2)
00 (0) = = 0:08293:
2(2)3=2
Concluding remarks.
1. The associate editor raised a question that we had also puzzled over ourselves: For
negative , is there a nice probabilistic interpretation of the analytic continuation of
E S ? In particular, one might entertain E (S ; < 1) and E (S j < 1) as
plausible candidates. In fact neither is correct. By inspection of (1), the analytic
continuation may be expressed as 1=(2E? S ) for 0. Dierentiating the
Wiener-Hopf factorization
[1 ? E (eiS ; < 1)][1 ? E (eiS? ; ? < 1)] = 1 ? E (eiX )
1
twice with respect to and setting = 0 shows that

1 = E (S ; < 1)
2E? S 1 + 2(?) + 2
for 0. This is easily seen to lie strictly between E (S ; < 1) and
E (S j < 1) as " 0.
2. After this work was done we learned of another occurrence of (1=2) in a related
problem; see Asmussen, Glynn, and Pitman (1995).
Acknowledgments. We are grateful to David Siegmund for passing along the

p
\interesting numerical curiosity" = ? (1=2)= 2, to Peter Jones for pointing us in the
direction of Hardy, and to Pavel Etingho for helpful discussions. Some of this work was
done at the Institute for Elementary Studies 1995 workshop in Pinecrest, California
(supported by NSF grant #DMS-9353149); we are grateful to Robin Pemantle for
inviting us.
REFERENCES
Asmussen, S. (1987). Applied Probability and Queues. Wiley, New York.
Asmussen, S., Glynn, P., and Pitman, J. (1995). Discretization error in simulation of
one-dimensional re ecting Brownian motion. Annals of Applied Probability 5
875{896.
Chang, J. T. (1992). On moments of the rst ladder height of random walks with small
drift. Annals of Applied Probability 2 714{738.
Cherno, H. (1965). Sequential test for the mean of a normal distribution IV (discrete
case). Ann. Math. Statist. 36 55{68.
Feller, W. (1971). An Introduction to Probability Theory and its Applications, Vol. II,
second edition. Wiley, New York.
Grandell, J. (1991). Aspects of Risk Theory. Springer, Berlin.
Hardy, G. H. (1905). A method for determining the behavior of certain classes of power
series near a singular point on the circle of convergence. Proc. London Math. Soc.
2 381{389.
Hogan, M. (1986). Comments on a problem of Cherno and Petkau. Ann. Probab. 14

1058{1063.
Lai, T. L. (1976). Asymptotic moments of random walks with applications to ladder

variables and renewal theory. Ann. Probab. 4 51{66.
Prabhu, N. U. (1980). Stochastic Storage Processes: Queues, Insurance Risk, and Dams.
Springer, New York.
Siegmund, D. (1979). Corrected diusion approximations in certain random walk

problems. Adv. Appl. Probab. 11 701{719.
Siegmund, D. (1985). Sequential Analysis: Tests and Condence Intervals. Springer, New
York.
Siegmund, D. (1986). Boundary crossing probabilities and statistical applications. Ann.

Statist. 14 361{404.
Titchmarsh, E. C. (1939). The Theory of Functions . Oxford University Press, London.
Whitt, W. (1974). Heavy trac limit theorems for queues: a survey. In Mathematical
Methods in Queueing Theory , A. B. Clarke, Ed., Lecture Notes in Economics and
Mathematical Systems 98 307{350. Springer, New York.
Woodroofe, M. (1982). Nonlinear Renewal Theory in Sequential Analysis. Society for

Industrial and Applied Mathematics, Philadelphia.
Dept. of Statistics, Yale University, Box 208290 Yale Station, New Haven, CT 06520
chang@stat.yale.edu
Dept. of Statistics, 367 Evans Hall, University of California, Berkeley, CA 94720 and
Mathematics Institute, The Hebrew University, Givat Ram, Jerusalem 91904, Israel
peres@stat.berkeley.edu

Ladder Heights, Gaussian Random Walks, and The Riemann Zeta Function

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ladder Heights, Gaussian Random Walks, and The Riemann Zeta Function

Uploaded by

Copyright:

Available Formats

Ladder Heights, Gaussian Random Walks,

and the Riemann Zeta Function

Yale University and University of California at Berkeley

Let fSn : n  0g be a random walk having normally distributed increments with

AMS 1991 subject classi cations. 60J15, 30B40.

For positive , the expected value of S depends on ; let us write it as E S to

Theorem 1.1 The expected rst ladder height E S may be extended to be an

will closely follow Hardy's.

a two-term asymptotic expansion of this series as z increases to 1 forms a cornerstone of

and expectation when the distribution of Xi is F . Under the assumptions that

(1979) showed that for a > 0,

normal family, these previous results gave

As for numerical computation, Cherno (1965) and Siegmund (1979) expressed  as an

This result involves relating the original problem to an associated problem

investigation is the asymptotic approximation of boundary crossing probabilities for

2 Analytic Continuation of ES

Di erentiating (5) with respect to , which can be justi ed by dominated convergence,

of the series around w = 0. This is done in Theorem 2.1 below.

Notation: It will be convenient to de ne c( ) = 2i=?(1 ? ). Since the Gamma

Theorem 2.1 For each 2 C, the function

Proof of Theorem 2.1. Removing the ray IR? = fx 2 IR : x  0g from the

log u = log juj + iarg(u);

(xy)z = xz yz if jarg(x) + arg(y)j < : (9)

For small positive , let C denote the contour

C = fx + i : ? ?1  x  0g + fei : =2    ?=2g + fx ? i : 0  x  ? ?1 g;

L = f?? + iy : ?  y  b ? 2g + fx + i(b ? 2) : ??  x  ag

this is the solid contour in Figure 1. Further, let

L = fx + i(b ? 2 ) : ?1 < x  ag + fa + yi : b ? 2  y  bg + fx + bi : a  x > ?1g

Lemma 2.2 (Hardy) Let H be as de ned in (7), and de ne

Proof. Let be arbitrary. Clearly H (w) is an analytic function of w for

combination of Fubini's and Morera's theorems, or section 2.83 of Titchmarsh (1939).

e?u ? e?w je?u j ? e

holds for Re(?u) > . Thus,

with the convergence being uniform over w 2 D . Therefore, 0

is analytic in w on the domain D , so that, by the arbitrary nature of D , (12) is analytic

on all of D, and hence so is I .

Next let Re( ) > 0. A change of variables in the de nition ?( ) = 1 x ? e?x dx

For w inside the closed contour C + L ,

Next observe that for Re( ) > 0,

convergence arguments show that for w 2 D n IR?,

#0 ?(1=) e ? ui

Thus, (15) follows from the statement

which holds for Re( ) > 0.

Therefore, by (12), (14), and (15),

= (e ?w ? 1)n for je?w ? 1j < je?u ? 1j:

Therefore, we may interchange sum and integral to obtain

for w 2 N . The convergence is uniform over w 2 N .

To evaluate the derivatives I k (0), write

de nes an entire function of , again by an easy combination of Fubini's and Morera's

using the function H .

H (k) (w) = I (k) (w) for w 2 D \ fRe(w) > 0g:

However, by Lemma 2.2, each derivative I k (w) is an analytic function of w for w 2 D.

Therefore, we may compute I k (0) as the limit

I (k) (0) = lim I (k) (w) = lim

(? 1)k c( ) n?( ?k) e?nw ? 2i( ? 1) ( ? k)w ?k?1

the entire complex plane, except for the (removable) singularity of c( ) ( ? k) at

Next we investigate the complete analytic continuation of the function  7! E S ; we

Theorem 2.4 For each integer J  0, the di erence

Proof. Start by writing

?(k + 1 ? ) = (?1)k ( ? 1)( ? 2) ( ? k)?(1 ? );

For J  0, use the de nition of J; (k) to write

j =1 k=0 k 2j j =1 2j

Let fSn : n 0g be a random walk having normally distributed increments with

AMS 1991 subject classications. 60J15, 30B40.

For positive , the expected value of S depends on ; let us write it as E S to

Theorem 1.1 The expected rst ladder height E S may be extended to be an

and expectation when the distribution of Xi is F . Under the assumptions that

As for numerical computation, Cherno (1965) and Siegmund (1979) expressed as an

2 Analytic Continuation of ES

Dierentiating (5) with respect to , which can be justied by dominated convergence,

Notation: It will be convenient to dene c() = 2i=?(1 ? ). Since the Gamma

Proof of Theorem 2.1. Removing the ray IR? = fx 2 IR : x 0g from the

(xy)z = xz yz if jarg(x) + arg(y)j < : (9)

For small positive , let C denote the contour

C = fx + i : ? ?1 x 0g + fei : =2 ?=2g + fx ? i : 0 x ? ?1 g;

L = f?? + iy : ? y b ? 2g + fx + i(b ? 2) : ?? x ag

L = fx + i(b ? 2 ) : ?1 < x ag + fa + yi : b ? 2 y bg + fx + bi : a x > ?1g

Lemma 2.2 (Hardy) Let H be as dened in (7), and dene

e?u ? e?w je?u j ? e

holds for Re(?u) > . Thus,

on all of D, and hence so is I.

Next let Re() > 0. A change of variables in the denition ?() = 1 x? e?x dx

For w inside the closed contour C + L ,

Next observe that for Re() > 0,

#0 ?(1=) e ? ui

which holds for Re() > 0.

To evaluate the derivatives Ik (0), write

denes an entire function of , again by an easy combination of Fubini's and Morera's

H(k) (w) = I(k) (w) for w 2 D \ fRe(w) > 0g:

However, by Lemma 2.2, each derivative Ik (w) is an analytic function of w for w 2 D.

Therefore, we may compute Ik (0) as the limit

I(k) (0) = lim I (k) (w) = lim

(? 1)k c() n?(?k) e?nw ? 2i( ? 1) ( ? k)w?k?1

the entire complex plane, except for the (removable) singularity of c() ( ? k) at

Next we investigate the complete analytic continuation of the function 7! E S ; we

Theorem 2.4 For each integer J 0, the dierence

For J 0, use the denition of J; (k) to write

j =1 k=0 k 2j j =1 2j

= ?2i(2) J; (k) sin[(k ? )=2]

J; (k) (J + 1)? k ? ( +1 Re( )) as k ! 1;

In summary, the function G(w) = c() P1 ? ?nw can be continued throughout

for each k, the dierence G ? B;k can be analytically continued to a neighborhood of

(and the attendant rays removed). In particular, E S is clearly continuable to a

P? f (b) < 1g = exp ?2 b ? p + O(e?rb)

P? f (b) < 1g = E (e?2S b ) = e?2b E (e?2Rb ):

jE (e? R ) ? ()j Ce?rb

1 ? E (e? S ) = 1 ? P? f < 1g = P f? = 1g = 1=(E );

For example, in a diusion normalization, which assumes that ! 0 and b ! 1

Corollary 3.2 If # 0 and b ! 1 in such a way that

P? f (b) < 1g exp ?2 b ? p 2

P? f (b) < 1g = exp ?2 b ? p

E0 (S4 ) given in the previous paragraph, is

twice with respect to and setting = 0 shows that

Siegmund, D. (1979). Corrected diusion approximations in certain random walk