HISTORY OF PROBABILITY AND STATISTICS

Kumpulan 4

AGNES NGU YU CHING

CLLY BIN GULIMIT FAZLI B. ALI KARIM FRANKIE AK PETER LANGAN FU SAI HOE LAVINA NAIR A/P MADAWAN NAIR

Ancient History

Astragali: six sided bones. Not symmetrical. the oldest known dice were excavated as part of a 5000-year-old backgammon set at the Burnt City, an archeological site in south-eastern Iran

The most popular dice game of the middle ages: hazard

Arabic al zhar means a die. brought to Europe by soldiers returning from the Crusades,

Rules much like modern-day craps.

Cards introduced 14th Primero: early form of poker.

Gaming

Gaming apparently existed in the earliest civilizations.

E.g., the talus a knucklebone or heel bone that can land in any of 4 different ways. Used for amusement.

French Society in the 1650s

Gambling

was popular and fashionable

Not

restricted by law

As

the games became more complicated and the stakes became larger there was a need for mathematical methods for computing chances.

Enter the Mathematicians

A well-known

gambler, the chevalier De Mere consulted Blaise Pascal in Paris about a some questions about some games of chance.

Pascal

began to correspond with his friend Pierre Fermat about these problems.

The Game of chance

In the mid-seventeenth century, a simple question directed to Blaise Pascal by a nobleman sparked the birth of probability theory, as we know it today. Chevalier de Mr gambled frequently to increase his wealth. He bet on a roll of a die that at least one 6 would appear during a total of four rolls. To find the probability of getting at least one 6 in four rolls, one would have to calculate the probability of getting one 6, two 6s, three 6s, and four 6s. Actually, it would be quicker to find the opposite of this event, called the complement. This is the probability of not getting any 6s. Since the probability of not rolling a 6 is five out of six for each of the four rolls, this probability is (5/6)4. To solve the game you need to subtract this probability from 1, which is 51.8 percent: 1 - (5/6)4

or 51.8 percent

Classical Probability
The

correspondence between Pascal and Fermat is the origin of the mathematical study of probability. method they developed is now called the classical approach to computing probabilities.

The

The

method: Suppose a game has n equally likely outcomes, of which m outcomes correspond to winning. Then the probability of winning is m/n.

Problems with the Classical Method

The

classical method requires a game to be broken down into equally likely outcomes: It is not always possible to do this. It is not always clear when possibilities are equally likely.

Experience
Another

method, known as the frequency method had also been used for some time. method consists of repeating a game a large number of times under the same conditions. The probability of winning is then approximately equal to the proportion of wins in the repeats. method was used by Pascal and Fermat to verify results obtained by the classical method.

This

This

Early Generalizations
James

Bernoulli proved that the frequency method and the classical method are consistent with one another in his book Ars Conjectandi in 1713.

Abraham De

Moivre provided many tools to make the classical method more useful, including the multiplication rule, in his book The Doctrine of Chances in 1718.

From Games to Science

Throughout

the 18th century, the application of probability moved from games of chance to scientific problems:

Mathematical theory of life insurance life tables.

Biological problems - what is the probability of being born female or male?

Applied Probability
Pierre-Simon

Laplace presented a mathematical theory of probability with an emphasis on scientific applications in his 1812 book Theorie Analytique des Probabilities. Laplace only considered the classical method, leaving no indication on how the method was to be applied to general problems.

Unfortunately,

Stagnation the Frustration

After

the publication of Laplaces book, the mathematical development of probability stagnated for many years. 1850, many mathematicians found the classical method to be unrealistic for general use and were attempting to redefine probability in terms of the frequency method. attempts were never fully accepted and the stagnation continued.

By

These

Axiomatic Development
Andrey

Kolmogorov developed the first rigorous approach to probability in his 1933 monograph Grundbegriffe der Wahrscheinlichkeitsrechnun. built up probability theory from fundamental axioms in a way comparable with Euclid's treatment of geometry.

He

Probability Today
Modern

research in probability theory is closely related to the mathematical field of measure theory.

Modern

innovators in the field include Patrick Billingsley (University of Chicago), Yuan Shih Chow (Columbia), Kai Lai Chung (Stanford), Samuel Karlin (Stanford), Rolf-Dieter Reiss, Sheldon Ross (Berkeley), Henry Teicher (Rutgers) and many more

History of Statistics

DEFINITION OF STATISTICS

By the 18th century, the term "statistics" designated the systematic collection of demographic and economic data by states. In the early 19th century, the meaning of "statistics" broadened to include the discipline concerned with the collection, summary, and analysis of data. In modern terms, "statistics" means both sets of collected information, as in national accounts and temperature records, and analytical work which requires statistical inference.

 The

relation between statistics and probability theory developed rather late, however. the 19th century, statistics increasingly used probability theory, whose initial results were found in the 17th and 18th centuries, particularly in the analysis of games of chance (gambling).

In

Antiquity
The

origin of descriptive statistics can be traced to ancient Rome and China,

the

Babylonians and Egyptians carried censuses in 4500-3000 B.C.

thousand years BC a population census took place in China registration of property was conducted in ancient Rome

Ancient Rome
Roman

Emperor Augustus (27 B.C.-17 A.D.) conducted surveys on births and deaths of the citizens of the empire as well as the amount of livestock each owned and the crops each harvested. order to use this information, the Romans had to develop methods of collecting, organizing, and summarizing data

In

Middle Ages
In

the Middle Ages, censuses of population, household goods and land were conducted. the 14th century, people began keeping records on births, deaths, and accidents in order to determine insurance rates

During

Time

Contributor

Contribution

Ancient Greece

Philosophers

Ideas - no quantitative analyses

Graunt, Petty 17th Century Pascal, Bernoulli

studied affairs of state, vital statistics of populations studied probability through games of chance, gambling

18th Century

Laplace, Gauss

normal curve, regression through study of astronomy

Quetelet 19th Century Galton

astronomer who first applied statistical analyses to human biology studied genetic variation in humans(used regression and correlation)

Pearson

studied natural selection using correlation, formed first academic department of statistics, Biometrika journal, helped develop the Chi Square analysis studied process of brewing, alerted the statistics community about problems with small sample sizes, developed Student's test evolutionary biologists - developed ANOVA, stressed the importance of experimental design biochemist studied pesticides, non-parametric equivalent of two-samples test economists who developed the non-parametric equivalent of the ANOVA psychologist who developed a non-parametric equivalent of the correlation coefficient statistician who developed another non-parametric equivalent the correlation coefficient statistician who developed multiple comparisons procedure biochemist who studied pesticides, developed multiple comparisons procedure for control groups agronomist who developed multiple comparisons procedure provided many advantages over calculations by hand or by

20th Century (early)

Gossett (Student)

Fisher

20th Century (later)

Wilcoxon Kruskal, Wallis Spearman Kendall Tukey Dunnett Keuls Computer Technology

Timeline of probability and statistics

17th century * 1654 - Blaise Pascal and Pierre de Fermat create the theory of probability, * 1693 - Edmund Halley prepares the first mortality tables statistically relating death rate to age, 18th century * 1724 - Abraham de Moivre studies mortality statistics and the foundation of the theory of annuities in "Annuities on Lives", * 1733 - Abraham de Moivre introduces the normal distribution to approximate the binomial distribution in probability, * 1761 - Thomas Bayes proves Bayes' theorem, 19th century * 1805 - Adrien-Marie Legendre introduces the method of least squares for fitting a curve to a given set of observations, 20th century * 1933 - Andrey Nikolaevich Kolmogorov publishes his book "Basic notions of the calculus of probability" ("Grundbegriffe der Wahrscheinlichkeitsrechnung") which contains an axiomatization of probability based on measure theory, * 1943 - Kenneth Levenberg proposes a method for nonlinear least squares fitting, * 1953 - Nicholas Metropolis introduces the idea of thermodynamic simulated annealing algorithms, * 1962 - Donald Marquardt proposes the Levenberg-Marquardt nonlinear least squares fitting algorithm,