- •Introduction
- •Basic concepts of probability theory
- •Classical definition of probability
- •Relative frequency
- •Geometric probabilities
- •Glossary
- •Exercises for Seminar 1
- •Exercises for Homework 1
- •Basic formulas of combinatorial analysis
- •Operations over events
- •Glossary
- •Exercises for Seminar 2
- •Exercises for Homework 2
- •Theorem of addition of probabilities of incompatible events
- •Complete group of events
- •Opposite events
- •Conditional probability
- •Theorem of multiplication of probabilities
- •Glossary
- •Exercises for Seminar 3
- •Exercises for Homework 3
- •Independent events
- •Where a is the appearance of at least one of the events a1, a2, …, An; .
- •Glossary
- •Exercises for Seminar 4
- •Exercises for Homework 4
- •Theorem of addition of probabilities of compatible events
- •Formula of total probability
- •Probability of hypotheses. Bayes’s formulas.
- •Glossary
- •Exercises for Seminar 5
- •Exercises for Homework 5
- •Repetition (recurrence) of trials. The Bernoulli formula
- •Local theorem of Laplace
- •Integral theorem of Laplace
- •Glossary
- •Exercises for Seminar 6
- •Exercises for Homework 6
- •Random variables. The law of distribution of a discrete random variable
- •A random variable is understood as a variable which as result of a trial takes one of the possible set of its values (which namely – it is not beforehand known).
- •Mathematical operations over random variables
- •(Mathematical) expectation of a discrete random variable
- •Dispersion of a discrete random variable
- •Glossary
- •Exercises for Seminar 7
- •Exercises for Homework 7
- •Distribution function of a random variable
- •Properties of a distribution function
- •Continuous random variables. Probability density
- •Properties of probability density
- •Glossary
- •Exercises for Seminar 8
- •Exercises for Homework 8
- •Basic laws of distribution of discrete random variables
- •1. Binomial law of distribution
- •2. The law of distribution of Poisson
- •3. Geometric distribution
- •4. Hypergeometric distribution
- •Glossary
- •Exercises for Seminar 9
- •Exercises for Homework 9
- •Basic laws of distribution of continuous random variables
- •1. The uniform law of distribution
- •2. Exponential law of distribution
- •3. Normal law of distribution
- •Glossary
- •Exercises for Seminar 10
- •Exercises for Homework 10
- •The law of large numbers and limit theorems
- •The central limit theorem
- •Glossary
- •Exercises for Seminar 11
- •Exercises for Homework 11
- •Mathematical statistics. Variation series and their characteristics
- •Numerical characteristics of variation series
- •Glossary
- •Exercises for Seminar 12
- •Exercises for Homework 12
- •Bases of the mathematical theory of sampling
- •Glossary
- •Exercises for Seminar 13
- •Exercises for Homework 13
- •Methods of finding of estimations
- •Notion of interval estimation
- •Glossary
- •Exercises for Seminar 14
- •Exercises for Homework 14
- •Testing of statistical hypotheses
- •Glossary
- •Exercises for Seminar 15
- •Exercises for Homework 15
- •Individual homeworks
- •Variant 1
- •Variant 2
- •Variant 3
- •Variant 4
- •Variant 5
- •Variant 6
- •Variant 7
- •Variant 8
- •Variant 9
- •Variant 10
- •Variant 11
- •Variant 12
- •Variant 13
- •Variant 14
- •Variant 15
- •Variant 16
- •Variant 17
- •Variant 18
- •Variant 19
- •Variant 20
- •Variant 21
- •Variant 22
- •Variant 23
- •Variant 24
- •Variant 25
- •Final exam trial tests (for self-checking)
- •Appendix
- •Values the functions and
- •List of the used books
- •Contents
3. Normal law of distribution
The normal law of distribution is most frequently met in practice. The main feature allocating it among other laws is that it is the limiting law to which other laws of distribution are approximated at rather frequently meeting typical conditions.
The normal distribution was introduced by the French mathematician Abraham DeMoivre in 1733 and was used by him to approximate probabilities associated with binomial random variables when the binomial parameter n is large. This result was later extended by Laplace and others and is now encompassed in a probability theory known as the central limit theorem, which is discussed later. The central limit theorem, one of the two most important results in probability theory, gives a theoretical base to the often noted empirical observation that, in practice, many random phenomena obey, at least approximately, a normal probability distribution. Some examples of this behavior are the height of a man, the velocity in any direction of a molecule in gas, and the error made in measuring a physical quantity.
A continuous random variable X has the normal law of distribution (Gauss law) with the parameters a and 2 if its probability density has the following form:
Theorem. The mathematical expectation of a random variable X distributed under the normal law is M(X) = a, and its dispersion D(X) = 2.
The normal law of distribution of a random variable with the parameters a = 0, 2 = 1, i.e. N(0; 1), is said to be standard or normalized.
A curve of the normal law of distribution is said to be normal or Gauss curve.
Theorem. The distribution function of a random variable X distributed under the normal law is expressed by the Laplace function (x):
where
.
Property 1. The probability of hit of a random variable X distributed under the normal law in the interval [x1; x2] is
where
Property 2. The probability that deviation of a random variable X distributed under the normal law from the mathematical expectation a doesn’t exceed the quantity > 0 (on absolute value) is equal to
where t = /.
The rule of three sigmas: If a random variable X has the normal distribution with the parameters a and 2, i.e. N(a; 2), then it is practically reliable that its values are in the interval (a – 3; a + 3), i.e.
The violation of «the rule of three sigmas», i.e. the deviation of normally distributed random variable X is more than on 3 (by absolute value), is an event which is practically impossible since its probability is rather small:
Example. Assuming that the height of men of a certain age group is a normally distributed random variable X with the parameters a = 173 and 2 = 36, find:
1. a) the probability density and the distribution function of the random variable X; b) parts of suits of the 4-th height (176-182) and the 3-rd height (170-176) which need to be provided in total amount of a factory for a given age group.
2. Formulate «the rule of three sigmas» for the random variable X.
Solution:
1. a) We have
b) The part of suits of the 4-th height (176 – 182 cm) in total amount of the factory is determined as probability:
since
The part of suits of the 3rd height (170 – 176 cm) can be determined analogously, but it is simply to make by the following way:
2. It is practically reliable that the height of men of the given age group is enclosed in boundaries from a – 3 = 173 – 3 6 = 155 to a + 3 = 173 + 3 6 = 191, i.e. 155 X 191 (cm).
Remark. DeMoivre, who used the normal distribution to approximate probabilities connected with coin tossing, called it the exponential bell-shaped curve. Its usefulness, however, became truly apparent only in 1809, when the famous German mathematician K.F. Gauss used it as an integral part of his approach to predicting the location of astronomic entities. As a result, it became common after this time to call the Gaussian distribution.
During the mid to late nineteenth century, however, most statisticians started to believe that the majority of data sets would have histograms conforming to the Gaussian bell-shaped form. Indeed, it came to be accepted that it was “normal” for any well-behaved data set to follow this curve. As a result, following the lead of the British statistician Karl Pearson, people began referring to the Gaussian curve by calling it simply the normal curve.
