Scribd
Upload
110 views

L1 Probability Theory

This document provides an introduction to probability theory, including definitions of key terms like experiment, sample space, events, unions and intersections of events, axioms of probability, and finite sample spaces. It outlines topics that will be covered like interpretations of probability, probability distributions, and results derived from the axioms. Tutorial sessions and exam structure are also summarized.

Uploaded by

fac_econ_presi
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
110 views

L1 Probability Theory

This document provides an introduction to probability theory, including definitions of key terms like experiment, sample space, events, unions and intersections of events, axioms of probability, and finite sample spaces. It outlines topics that will be covered like interpretations of probability, probability distributions, and results derived from the axioms. Tutorial sessions and exam structure are also summarized.

Uploaded by

fac_econ_presi
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Probability Theory

003 Introductory Econometrics


Deepti Goel Part I Sunil Kanwar Part II

December 2014

Probability Theory .

Outline

Probability Theory
Experiments and Events Random Variables Distributions

Statistical Inference
Maximum Likelihood Estimation Sampling distributions Testing Hypotheses (if time permits)

Probability Theory .

Outline

Main Text: Probability and Statistics by Morris H. Degroot and Mark J. Schervish (Third edition). Available in RTL. Concentrate on odd numbered problems as these have solutions in the text. These would be su cient practice for this course.

Probability Theory .

Evaluation

Internal Assessment: 50%


Part 1: at least 20% Midterm for Part 1 will be on March 3, 2014 at 9:10 am Part 2: 25%

Final Exam: Part I and Part II , each constituting 25%

Probability Theory .

Tutorials, Lab sessions and Contact hours

Tutorials
One tutorial session per week: details to be announced soon. Check CDE and LT notice boards. Tutors (Atika Gupta and Reetika Garg) will discuss problem sets; these will not be graded, but would be good practice for nal exam.

Computer Lab Sessions


To be decided

Contact timing: You can meet me after class to schedule an appointment.

Probability Theory .

Interpretations of Probability
Probability is a number attached to the likelihood of something occuring. How can these numbers be attached? Frequency Interpretation: The relative frequency with which the outcome would occur where the process repeated a large number of times under similar conditions Classical Interpretation: If there are n possible outcomes of a process and if each of them is equally likely, then the probability of each outcome is 1/n. Subjective Interpretation: The probability of an outcome of a process depends on the person s judgement about the likelihood of it occuring. This inturn depends on the person s beliefs and information. Dierent people may have dierent subjective probabilities for the same outcome.

Probability Theory .

Interpretations of Probability

Probability theory gives scientic structure to situations involving uncertainty and is meaningful no matter what interpretation of probability one uses

Probability Theory .

Probability Theory: denitions

Experiment: A process whose outcome is not known in advance with certainty. All possible outcomes can listed out before the process is performed. Sample Space S : Set or collection of all possible outcomes of an experiment. Event: A dened subset in S . An event is always spoken of in the context of a particular experiment and the corresponding S Empty set : It is a set that contains no outcomes. It is dened to be a subset of every set.

Probability Theory .

Probability Theory: denitions


If A and B are any two events, the union of A and B is dened as the event containing all outcomes that belong to A alone, to B alone, or to both A and B , ( A [ B )

If A and B are any two events, the intersection of A and B is dened as the event containing all outcomes that belong to A and to B , ( A \ B ) Complement of an event A, is that subset of S that contains all outcomes that are not in A, (AC ) A1 , A2 , ... are mutually exclusive events or disjoint events, if no two of them have an outcome in common. If A1 , A2 , ... are events such that when we consider the set consisting of all outcomes that are contained in at least one of A1 , A2 , ... we get S , then A1 , A2 , ... are called exhaustive events.

Probability Theory .

Axiomatic Denition of Probability


Let F be the collection of all events in sample space S . A probability distribution is a function, Pr : F ! R, that assigns to each event A in F ,a number, Pr (A), called probability of A, that satises the following axioms:
The probability of every event A in F must be nonnegative For every event A, Pr(A) > 0, If an event is certain to occur then its probability is 1 Pr(S ) = 1 For every innite sequence of disjoint events A1 , A2 , ... Pr([i =1 Ai ) =

i =1

Pr(Ai )

Probability Theory .

Results derived from axioms, Proofs in Text


Pr() = 0 For every nite sequence of n disjoint events A1 , A2 , ...An , Pr([n i = 1 Ai ) = For every event A, Pr (Ac ) = 1 If A is a subset of B ,then Pr(A) For every event A, 0 Pr(A) 1. For every two events A and B , Pr(A [ B ) = Pr(A) + Pr(B ) Pr(A \ B )
n

i =1

Pr(Ai )
Pr(A) Pr(B )

Probability Theory .

Finite Sample Spaces


Sample space S of an experiment is nite if it contains only nite number of outcomes, S = fs1 , s2 , ...sn g

We can specify a probability distribution on a nite sample space S by assigning a probability pi to each element si in S . These probabilities must satisfy two conditions:
pi > 0 8 i n i = 1 pi = 1

The probability of any event A is obtained by adding the probabilities pi of all outcomes si belonging to A A sample space S conatining n outcomes s1 , s2 , ...sn is called a simple sample space if the probability assigned to each outcome is 1/n.
If an event A contains exactly m outcomes then, Pr (A) = m /n

Probability Theory .

Finite Sample Spaces

The sample space for the same experiment can be dened in a number of ways. Consider tossing of three coins. Suppose we are interested in the number of heads obtained. Sample space can be dened as fTTT , TTH , THT , HTT , HHT , HTH , THH , HHH g or as f0, 1, 2, 3g. The rst is a simple sample space, the second is not. We can arrive at total number of sample points in a sample space by listing all possible outcomes. Listing all possible outcomes can take a long time. Counting methods make things easier.

You might also like