In mathematics, tightness is a concept in measure theory. The intuitive idea is that a given collection of measures does not "escape to infinity".
Definitions
[edit]Let be a Hausdorff space, and let be a σ-algebra on that contains the topology . (Thus, every open subset of is a measurable set and is at least as fine as the Borel σ-algebra on .) Let be a collection of (possibly signed or complex) measures defined on . The collection is called tight (or sometimes uniformly tight) if, for any , there is a compact subset of such that, for all measures ,
where is the total variation measure of . Very often, the measures in question are probability measures, so the last part can be written as
If a tight collection consists of a single measure , then (depending upon the author) may either be said to be a tight measure or to be an inner regular measure.
If is an -valued random variable whose probability distribution on is a tight measure then is said to be a separable random variable or a Radon random variable.
Another equivalent criterion of the tightness of a collection is sequentially weakly compact. We say the family of probability measures is sequentially weakly compact if for every sequence from the family, there is a subsequence of measures that converges weakly to some probability measure . It can be shown that a family of measure is tight if and only if it is sequentially weakly compact.
Examples
[edit]Compact spaces
[edit]If is a metrizable compact space, then every collection of (possibly complex) measures on is tight. This is not necessarily so for non-metrisable compact spaces. If we take with its order topology, then there exists a measure on it that is not inner regular. Therefore, the singleton is not tight.
Polish spaces
[edit]If is a Polish space, then every probability measure on is tight. Furthermore, by Prokhorov's theorem, a collection of probability measures on is tight if and only if it is precompact in the topology of weak convergence.
A collection of point masses
[edit]Consider the real line with its usual Borel topology. Let denote the Dirac measure, a unit mass at the point in . The collection
is not tight, since the compact subsets of are precisely the closed and bounded subsets, and any such set, since it is bounded, has -measure zero for large enough . On the other hand, the collection
is tight: the compact interval will work as for any . In general, a collection of Dirac delta measures on is tight if, and only if, the collection of their supports is bounded.
A collection of Gaussian measures
[edit]Consider -dimensional Euclidean space with its usual Borel topology and σ-algebra. Consider a collection of Gaussian measures
where the measure has expected value (mean) and covariance matrix . Then the collection is tight if, and only if, the collections and are both bounded.
Tightness and convergence
[edit]Tightness is often a necessary criterion for proving the weak convergence of a sequence of probability measures, especially when the measure space has infinite dimension. See
- Finite-dimensional distribution
- Prokhorov's theorem
- Lévy–Prokhorov metric
- Weak convergence of measures
- Tightness in classical Wiener space
- Tightness in Skorokhod space
Exponential tightness
[edit]A strengthening of tightness is the concept of exponential tightness, which has applications in large deviations theory. A family of probability measures on a Hausdorff topological space is said to be exponentially tight if, for any , there is a compact subset of such that
References
[edit]- Billingsley, Patrick (1995). Probability and Measure. New York, NY: John Wiley & Sons, Inc. ISBN 0-471-00710-2.
- Billingsley, Patrick (1999). Convergence of Probability Measures. New York, NY: John Wiley & Sons, Inc. ISBN 0-471-19745-9.
- Ledoux, Michel; Talagrand, Michel (1991). Probability in Banach spaces. Berlin: Springer-Verlag. pp. xii+480. ISBN 3-540-52013-9. MR1102015 (See chapter 2)