Distribution dégénérée
A distribution dégénérée is a special type of probability distribution where all the probability mass is concentrated at a single point or value. This means that there is no variability in the outcomes; the random variable takes on only one specific value with a probability of 1.
Mathematically, if X is a random variable with a degenerate distribution, then there exists a constant ‘c’ such that:
X = c with probability 1.
This can be visualized as a distribution that has a spike at the value ‘c’ on the number line, with the probability density function (PDF) being zero everywhere else. The fonction de distribution cumulative (CDF) for a degenerate distribution jumps from 0 to 1 at the point ‘c’.
Les distributions dégénérées sont souvent utilisées en théorie statistics and probability to simplify problems, particularly when dealing with deterministic outcomes or fixed values. For example, in a scenario where an experiment always yields the same result, the outcome can be modeled using a degenerate distribution.
While degenerate distributions may seem trivial, they serve important roles in various méthodes statistiques, such as inférence bayésienne, where prior beliefs can be represented as degenerate distributions centered around a specific hypothesis.