MCS 341 EXAM 4
- When: Available November 22. Due November 24
- What:
An examination consisting of problems related to
Chapter 5, Multivariate Probability Distributions
and supplementary class notes.
- How: Take-home and closed-book,
but with one new 3"-by-5" note card allowed.
(You may also use your note cards from the previous exams.)
TOPICS
- Bivariate and multivariate probability distributions
- The joint distribution function
for jointly distributed random variables
Y1, ..., Yn,
i.e., r.v.s defined on the same probability space,
F(y1, ..., yn) =
P(Y1<y1, ...,
Yn<yn),
is always defined.
- Jointly distributed "continuous" random variables are
those whose joint distribution function is continuous.
- Some, but not all, jointly distributed continuous r.v.s
have a joint probability density function.
Their joint distribution function and other probabilities may
be calculated by integrating the joint pdf.
- Jointly distributed discrete random variables have a
joint probability function.
- Marginal and conditional probability distributions
- The marginal distributions are found by summation
(discrete case) or integration (pdf case) of the joint
probability function or joint pdf.
- The conditional probability function or pdf is found
by dividing the joint probability function or pdf by the
appropriate marginal function.
- Independent random variables
- Definition
- Tests for independence, including Theorem 5.5
- Note the clarifications given in class.
- The expected value of a function of random variables:
the Law of the Unconscious Statistician
- Special theorems, especially linearity
- The covariance of two random variables
- Definition and computing formulas
- Correlation coefficient
- Properties of covariance and correlation.
- Relationship between uncorrelatedness and independence.
- Expected value, variance, and covariance of linear combinations
of random variables
- The multinomial probability distribution
- Joint probability function
- Expected values, variances, and covariances
- The bivariate and multivariate normal distributions will not
be covered on this exam.
- Conditional expectations