Title: Laplace, Pierre Simon de (1749-1827)
1Laplace, Pierre Simon de (1749-1827)
2Agenda
- Biography
- History of The Central limit Theorem (CLT)
- Derivation of the CLT
- First Version of the CLT
- CLT for the Binomial Distribution
- Laplace and Bayesian ideas
- 4th and 5th Principles
- 6th Principle
- 7th Principle
- References
3Biography
- Pierre Simon Laplace was born in Normandy on
March 23, 1749, and died at Paris on March 5,
1827 - French scientist, mathematician and astronomer
established mathematically the stability of the
Solar system and its origin - without a divine
intervention - Professor of mathematics in the École militaire
of Paris at the age of 19.
4Biography (Contd)
- Under Napoleon, Laplace was a member, then
chancellor, of the Senate, and received the
Legion of Honour in 1805. Appointed Minister of
the Interior, in 1799. Removed from office by
Napoleon after six weeks only!! - Named a Marquis in 1817 after the restoration of
the Bourbons - Main publications
- Mécanique céleste (1771, 1787)
- Théorie analytique des probabilités 1812 first
edition dedicated to Napoleon
5History Of Central Limit Theorem From De Moivre
to Laplace
- De Moivre investigated the limits of the binomial
distribution as the number of trials increases
without bound and found that the function
exp(-x2) came up in connection with this problem. - The formulation of the normal distribution,
(1/v2)exp(-x2/2), came with Thomas Simpson.
6History Of CLT (Contd)
- This idea was was expanded upon by the German
mathematician Carl Friedrich Gauss who then
developed the principle of least squares. - Independently, the French mathematicians Pierre
Simon de Laplace and Legendre also developed
these ideas. It was with Laplace's work that the
first inklings of the Central Limit Theorem
appeared. - In France, the normal distribution is known as
Laplacian Distribution while in Germany it is
known as Gaussian.
7Derivation of the CLT
- Initial Work Laplace was calculating the
probability distribution of the sum of meteor
inclination angles. He assumed that all the
angles were r.vs following a triangular
distribution between 0 and 90 degrees - Problems
- The deviation between the arithmetic mean which
was inflicted with observational errors, and the
theoretical value - The exact calculation was not achievable due to
the considerable amount of celestrial bodies - Solution Find an approximation !!
8First Version of the CLT
- Laplace introduced the m.g.f,
which is known as Laplace Transform of
f - He then introduced the Characteristic function
- If is a sample of i.i.d.
obs., - Assume that we have a discrete r.v. x, that takes
on the values - m, -m1,,0,,m-1,m with prob. p-m ,,pm.
- Let Sn be the sum of the n possible errors.
9(No Transcript)
10CLT for the Binomial Distribution
11Final Note on The Proof of The CLT
- It was Lyapunov's analysis that led to the modern
characteristic function approach to the Central
Limit Theorem. - Where
12Laplace Bayesian ideasOverview
Philosophical essay on probabilities by Laplace
- General principles on probability
- Expectation
- Analytical methods
- Applications
134th 5th principles conditional marginal give
the joint
Here the question posed by some philosophers
concerning the influence of the past on the
probability of the future, presents itself
146th principle discrete Bayes theorem
Fundamental principle of that branch of the
analysis of chance that consists of reasoning a
posteriori from events to causes
157th principle probability of future based on
observations
() the correct way of relating past events to
the probability of causes and of future events
()
16References
- Laplace, Pierre Simon De. Philosophical essay on
probabilities. - Weatherburn, C.E. A First Course in Mathematical
Statistics.