Probability theory

Author: Michal Wardzynski
Supervisor : Igor Ohirko
Facility: Kazimierz Pułaski University of Technology and Humanities in Radom

Probability theory is the branch of mathematics concerned with probability. Although there are several different probability interpretations, probability theory treats the concept in a rigorous mathematical manner by expressing it through a set of axioms. Typically these axioms formalise probability in terms of a probability space, which assigns a measure taking values between 0 and 1, termed the probability measure, to a set of outcomes called the sample space. Any specified subset of these outcomes is called an event.

Central subjects in probability theory include discrete and continuous random variables, probability distributions, and stochastic processes, which provide mathematical abstractions of non-deterministic or uncertain processes or measured quantities that may either be single occurrences or evolve over time in a random fashion.

Although it is not possible to perfectly predict random events, much can be said about their behaviour. Two major results in probability theory describing such behaviour are the law of large numbers and the central limit theorem.

As a mathematical foundation for statistics, probability theory is essential to many human activities that involve quantitative analysis of data. Methods of probability theory also apply to descriptions of complex systems given only partial knowledge of their state, as in statistical mechanics. A great discovery of twentieth century physics was the probabilistic nature of physical phenomena at atomic scales, described in quantum mechanics.

History

The mathematical theory of probability has its roots in attempts to analyze games of chance by Gerolamo Cardano in the sixteenth century, and by Pierre de Fermat and Blaise Pascal in the seventeenth century (for example the „problem of points”). Christiaan Huygens published a book on the subject in 1657 and in the 19th century, Pierre Laplace completed what is today considered the classic interpretation.

Initially, probability theory mainly considered discrete events, and its methods were mainly combinatorial. Eventually, analytical considerations compelled the incorporation of continuous variables into the theory.

This culminated in modern probability theory, on foundations laid by Andrey Nikolaevich Kolmogorov. Kolmogorov combined the notion of sample space, introduced by Richard von Mises, and measure theory and presented his axiom system for probability theory in 1933. This became the mostly undisputed axiomatic basis for modern probability theory; but, alternatives exist, such as the adoption of finite rather than countable additivity by Bruno de Finetti.

Treatment

Most introductions to probability theory treat discrete probability distributions and continuous probability distributions separately. The more mathematically advanced measure theory-based treatment of probability covers the discrete, continuous, a mix of the two, and more.

Motivation

Consider an experiment that can produce a number of outcomes. The set of all outcomes is called the sample space of the experiment. The power set of the sample space (or equivalently, the event space) is formed by considering all different collections of possible results. For example, rolling an honest die produces one of six possible results. One collection of possible results corresponds to getting an odd number. Thus, the subset {1,3,5} is an element of the power set of the sample space of die rolls. These collections are called events. In this case, {1,3,5} is the event that the die falls on some odd number. If the results that actually occur fall in a given event, that event is said to have occurred.

Probability is a way of assigning every „event” a value between zero and one, with the requirement that the event made up of all possible results (in our example, the event {1,2,3,4,5,6}) be assigned a value of one. To qualify as a probability distribution, the assignment of values must satisfy the requirement that if you look at a collection of mutually exclusive events (events that contain no common results, e.g., the events {1,6}, {3}, and {2,4} are all mutually exclusive), the probability that any of these events occurs is given by the sum of the probabilities of the events.

The probability that any one of the events {1,6}, {3}, or {2,4} will occur is 5/6. This is the same as saying that the probability of event {1,2,3,4,6} is 5/6. This event encompasses the possibility of any number except five being rolled. The mutually exclusive event {5} has a probability of 1/6, and the event {1,2,3,4,5,6} has a probability of 1, that is, absolute certainty.

Discrete probability distributions

The Poisson distribution, a discrete probability distribution.

Discrete probability theory deals with events that occur in countablesample spaces.

Examples: Throwing dice, experiments with decks of cards, random walk, and tossing coins

Classical definition: Initially the probability of an event to occur was defined as number of cases favorable for the event, over the number of total outcomes possible in an equiprobable sample space: see Classical definition of probability.

For example, if the event is „occurrence of an even number when a die is rolled”, the probability is given by {\tfrac {3}{6}}={\tfrac {1}{2}}, since 3 faces out of the 6 have even numbers and each face has the same probability of appearing.

Modern definition: The modern definition starts with a finite or countable set called the sample space, which relates to the set of all possible outcomes in classical sense, denoted by \Omega . It is then assumed that for each element x\in \Omega \,, an intrinsic „probability” value f(x)\, is attached, which satisfies the following properties:

  1. {f(x)\in [0,1]{\mbox{ for all }}x\in \Omega \,;
  2. \sum _{x\in \Omega }f(x)=1\,.

That is, the probability function f(x) lies between zero and one for every value of x in the sample space Ω, and the sum of f(x) over all values x in the sample space Ω is equal to 1. An event is defined as any subset E\, of the sample space \Omega \,. The probability of the event E\, is defined as

P(E)=\sum _{x\in E}f(x)\,.

So, the probability of the entire sample space is 1, and the probability of the null event is 0.

The function f(x)\, mapping a point in the sample space to the „probability” value is called a probability mass functionabbreviated as pmf. The modern definition does not try to answer how probability mass functions are obtained; instead it builds a theory that assumes their existence

Continuous probability distributions

The normal distribution, a continuous probability distribution.

Continuous probability theory deals with events that occur in a continuous sample space.

Classical definition: The classical definition breaks down when confronted with the continuous case. See Bertrand’s paradox.

Modern definition: If the outcome space of a random variable X is the set of real numbers (\mathbb {R} ) or a subset thereof, then a function called the cumulative distribution function (or cdfF\, exists, defined by F(x)=P(X\leq x)\,. That is, F(x) returns the probability that X will be less than or equal to x.

The cdf necessarily satisfies the following properties.

  1. F\, is a monotonically non-decreasing, right-continuous function;
  2. \lim _{x\rightarrow -\infty }F(x)=0\,;
  3. \lim _{x\rightarrow \infty }F(x)=1\,.

If F\, is absolutely continuous, i.e., its derivative exists and integrating the derivative gives us the cdf back again, then the random variable X is said to have a probability density function or pdf or simply density f(x)={\frac {dF(x)}{dx}}\,.

For a set E\subseteq \mathbb {R} , the probability of the random variable X being in E\, is

P(X\in E)=\int _{x\in E}dF(x)\,.

In case the probability density function exists, this can be written as

P(X\in E)=\int _{x\in E}f(x)\,dx\,.

Whereas the pdf exists only for continuous random variables, the cdf exists for all random variables (including discrete random variables) that take values in \mathbb {R} \,.

These concepts can be generalized for multidimensional cases on \mathbb {R} ^{n} and other continuous sample spaces.

Measure-theoretic probability theory

The raison d’être of the measure-theoretic treatment of probability is that it unifies the discrete and the continuous cases, and makes the difference a question of which measure is used. Furthermore, it covers distributions that are neither discrete nor continuous nor mixtures of the two.

An example of such distributions could be a mix of discrete and continuous distributions—for example, a random variable that is 0 with probability 1/2, and takes a random value from a normal distribution with probability 1/2. It can still be studied to some extent by considering it to have a pdf of (\delta [x]+\varphi (x))/2, where \delta [x] is the Dirac delta function.

Other distributions may not even be a mix, for example, the Cantor distribution has no positive probability for any single point, neither does it have a density. The modern approach to probability theory solves these problems using measure theory to define the probability space:

Given any set \Omega \, (also called sample space) and a σ-algebra {\mathcal {F}}\, on it, a measure P\, defined on {\mathcal {F}}\, is called a probability measure if P(\Omega )=1.\,

If {\mathcal {F}}\, is the Borel σ-algebra on the set of real numbers, then there is a unique probability measure on {\mathcal {F}}\, for any cdf, and vice versa. The measure corresponding to a cdf is said to be induced by the cdf. This measure coincides with the pmf for discrete variables and pdf for continuous variables, making the measure-theoretic approach free of fallacies.

The probability of a set E\, in the σ-algebra {\mathcal {F}}\, is defined as

P(E)=\int _{\omega \in E}\mu _{F}(d\omega )\,

where the integration is with respect to the measure \mu _{F}\, induced by F\,.

Along with providing better understanding and unification of discrete and continuous probabilities, measure-theoretic treatment also allows us to work on probabilities outside \mathbb {R} ^{n}, as in the theory of stochastic processes. For example, to study Brownian motion, probability is defined on a space of functions.

When it’s convenient to work with a dominating measure, the Radon-Nikodym theorem is used to define a density as the Radon-Nikodym derivative of the probability distribution of interest with respect to this dominating measure. Discrete densities are usually defined as this derivative with respect to a counting measure over the set of all possible outcomes. Densities for absolutely continuous distributions are usually defined as this derivative with respect to the Lebesgue measure. If a theorem can be proved in this general setting, it holds for both discrete and continuous distributions as well as others; separate proofs are not required for discrete and continuous distributions.

 

Reklamy

Skomentuj

Wprowadź swoje dane lub kliknij jedną z tych ikon, aby się zalogować:

Logo WordPress.com

Komentujesz korzystając z konta WordPress.com. Wyloguj /  Zmień )

Zdjęcie na Google

Komentujesz korzystając z konta Google. Wyloguj /  Zmień )

Zdjęcie z Twittera

Komentujesz korzystając z konta Twitter. Wyloguj /  Zmień )

Zdjęcie na Facebooku

Komentujesz korzystając z konta Facebook. Wyloguj /  Zmień )

Połączenie z %s