Infinite sum over a topological monoid #
This sum is known as unconditionally convergent, as it sums to the same value under all possible permutations. For Euclidean spaces (finite dimensional Banach spaces) this is equivalent to absolute convergence.
Note: There are summable sequences which are not unconditionally convergent! The other way holds
generally, see has_sum.tendsto_sum_nat
.
References #
- Bourbaki: General Topology (1995), Chapter 3 §5 (Infinite sums in commutative groups)
Infinite sum on a topological monoid
The at_top
filter on finset β
is the limit of all finite sets towards the entire type. So we sum
up bigger and bigger sets. This sum operation is invariant under reordering. In particular,
the function ℕ → ℝ
sending n
to (-1)^n / (n+1)
does not have a
sum for this definition, but a series which is absolutely convergent will have the correct sum.
This is based on Mario Carneiro's
infinite sum df-tsms
in Metamath.
For the definition or many statements, α
does not need to be a topological monoid. We only add
this assumption later, for the lemmas where it is relevant.
Equations
- has_sum f a = filter.tendsto (λ (s : finset β), ∑ (b : β) in s, f b) filter.at_top (𝓝 a)
summable f
means that f
has some (infinite) sum. Use tsum
to get the value.
∑' i, f i
is the sum of f
it exists, or 0 otherwise
Constant zero function has sum 0
If a function f
vanishes outside of a finite set s
, then it has_sum
∑ b in s, f b
.
If f : ℕ → α
has sum a
, then the partial sums ∑_{i=0}^{n-1} f i
converge to a
.
If a series f
on β × γ
has sum a
and for each b
the restriction of f
to {b} × γ
has sum g b
, then the series g
has sum a
.
You can compute a sum over an encodably type by summing over the natural numbers and taking a supremum. This is useful for outer measures.
tsum_supr_decode₂
specialized to the complete lattice of sets.
Some properties about measure-like functions.
These could also be functions defined on complete sublattices of sets, with the property
that they are countably sub-additive.
R
will probably be instantiated with (≤)
in all applications.
If a function is countably sub-additive then it is sub-additive on encodable types
If a function is countably sub-additive then it is sub-additive on finite sets
If a function is countably sub-additive then it is binary sub-additive
Let f : β → α
be a sequence with summable series and let b ∈ β
be an index.
Lemma tsum_ite_eq_extract
writes Σ f n
as the sum of f b
plus the series of the
remaining terms.
Sums on subtypes #
If s
is a finset of α
, we show that the summability of f
in the whole space and on the subtype
univ - s
are equivalent, and relate their sums. For a function defined on ℕ
, we deduce the
formula (∑ i in range k, f i) + (∑' i, f (i + k)) = (∑' i, f i)
, in sum_add_tsum_nat_add
.
For f : ℕ → α
, then ∑' k, f (k + i)
tends to zero. This does not require a summability
assumption on f
, as otherwise all sums are zero.
The Cauchy criterion for infinite sums, also known as the Cauchy convergence test
The sum over the complement of a finset tends to 0
when the finset grows to cover the whole
space. This does not need a summability assumption, as otherwise all sums are zero.
Series divergence test: if f
is a convergent series, then f x
tends to zero along
cofinite
.
For infinite sums taking values in a linearly ordered monoid, the existence of a least upper bound for the finite sums is a criterion for summability.
This criterion is useful when applied in a linearly ordered monoid which is also a complete or
conditionally complete linear order, such as ℝ
, ℝ≥0
, ℝ≥0∞
, because it is then easy to check
the existence of a least upper bound.
Alias of summable_abs_iff
.
Alias of summable_abs_iff
.
If the extended distance between consecutive points of a sequence is estimated
by a summable series of nnreal
s, then the original sequence is a Cauchy sequence.
If the distance between consecutive points of a sequence is estimated by a summable series, then the original sequence is a Cauchy sequence.
Multipliying two infinite sums #
In this section, we prove various results about (∑' x : β, f x) * (∑' y : γ, g y)
. Note that we
always assume that the family λ x : β × γ, f x.1 * g x.2
is summable, since there is no way to
deduce this from the summmabilities of f
and g
in general, but if you are working in a normed
space, you may want to use the analogous lemmas in analysis/normed_space/basic
(e.g tsum_mul_tsum_of_summable_norm
).
We first establish results about arbitrary index types, β
and γ
, and then we specialize to
β = γ = ℕ
to prove the Cauchy product formula (see tsum_mul_tsum_eq_tsum_sum_antidiagonal
).
Arbitrary index types #
Product of two infinites sums indexed by arbitrary types.
See also tsum_mul_tsum_of_summable_norm
if f
and g
are abolutely summable.
ℕ
-indexed families (Cauchy product) #
We prove two versions of the Cauchy product formula. The first one is
tsum_mul_tsum_eq_tsum_sum_range
, where the n
-th term is a sum over finset.range (n+1)
involving nat
substraction.
In order to avoid nat
substraction, we also provide tsum_mul_tsum_eq_tsum_sum_antidiagonal
,
where the n
-th term is a sum over all pairs (k, l)
such that k+l=n
, which corresponds to the
finset
finset.nat.antidiagonal n
The Cauchy product formula for the product of two infinites sums indexed by ℕ
,
expressed by summing on finset.nat.antidiagonal
.
See also tsum_mul_tsum_eq_tsum_sum_antidiagonal_of_summable_norm
if f
and g
are absolutely summable.
The Cauchy product formula for the product of two infinites sums indexed by ℕ
,
expressed by summing on finset.range
.
See also tsum_mul_tsum_eq_tsum_sum_range_of_summable_norm
if f
and g
are absolutely summable.