Main Index
Mathematical analysis
Infinite series and products
Infinite products
Subject Index
comment on the page
Infinite product appeared for the first time [1] in the work of Viète [2] , p. 400, who found
the product involving ![]()

Almost simultaneously in 1656 J.Wallis [3] , p. 468, found the expression
![]()
Euler later found many infinite product expressions, but the rigorous convergence theory of infinite products was initiated by Cauchy.
An infinite product is an expression of the form
![]()
This infinite product is called convergent
Otherwise the product is said to diverge. If both conditions are satisfied the product is assigned the value
. Because of the first condition it is often used the notation
indicating that it is sufficient to consider the products only from some index onwards. For the sake of simplicity, in what follows we shall always start with
.
There follows from the definition that a convergent infinite product vanishes if and only if on of its factors vanishes.
Theorem: If an infinite product converges then
..
Previous result motivates to denote the factors of an infinite product in the form
, where if the product is convergent we can suppose that
, that is to consider the products of the type
| (1) |
Theorem (Cauchy-Bolzano criterion): The infinite product
converges if and only if for every
there exists
such that for every
and every
we have
![]()
Theorem: The infinite product
converges if and only if the series
converges for some
(here we take the principal branch of the logarithm).
Proof. Without loss of generality we can suppose that
. Let
, and
be the
th partial sum of the infinite series and
the
th product of the infinite product, respectively. Then
.
Suppose that the series converges and let
. Then by continuity
![]()
and the Theorem is proved in one direction.
In the opposite direction the problem is caused by the fact that
does not converge to
but to one of its branches.
For every
there is an integer
such that
![]()
Taking the difference we get
![]()
and thus for the arguments we have
![]()
Since
, l
as
. Thus the first two terms on the right hand side are approaching each other, while the absolute values of the last term is at most
. Consequently,
for all sufficiently large
, and the series converges.
An infinite product
is called absolutely convergent if the product
is convergent.
Theorem: The product
converges absolutely if and only if the series
converges absolutely.
Theorem: If the product
converges then also the product
converges.
Theorem: The product
converges absolutely if the series
converges and the series
converges absolutely.
Note that if
and if
converges then clearly it converges absolutely, but this is not true for complex
.
Theorem: If the series
converges absolutely and if for
we have
, then
![]()
exists and is finite and non-vanishing.
Note that the previous result does not depend on the convergence behavior of the series
.
To the proof note
that
, that is
![]()
Since the
’s are bounded, the exponent in the second factor converges, and the Theorem follows.
Corollary: If the series
converges absolutely, the series
and the product
have the same convergence behavior.
If in (1) the numbers
are non-negative, the product is called with non-negative terms.
Theorem: The product
with
converges if and only if the series
converges.
Theorem: The product
with
converges if and only if the series
converges.
The products
and
are convergent for
and divergent for
.
If
then
monotonically decreases and consequently converges. If in addition
diverges the
, and the product
diverges to 0.
| [1] | Knopp, K. (1996). Theorie und Anwendung der unendlichen Reihen. (Theory and applications of infinite series).6th ed.. Berlin: Springer Verlag. |
| [2] | Viète, F. (1646). Opera Mathematica. Leyden; reprinted London, 1970 |
| [3] | Wallis, J. (1695). Opera I. Osord |
Cite this web-page as:
Štefan Porubský: Infinite products.