ST2004 - Applied Probability I 2017/2018 Module Code: ST2004 Module Title: Applied Probability I (module descriptor) Credit Value: 5 Cohorts: MSISS, CSL(JS) Week range: 05-10, 12-16 Semester: S1 (Michaelmas) Total hours: 33
Additional material will be provided when needed.
26/09/17: Monte Carlo approach, Empirical Law of Large Numbers, Birthday problem (via both simulation and combinatorics). Textbook: Section 2 (pp. 18-25), Section 3 (pp. 75-80, 3.1.4 excluded). --- 27/09/17: True and pseudo random number generation, Multiplicative congruential method, Random generation from an interval, Random generation from integers, Generation of random permutations. Textbook: Section 2.8 (pp. 52-59), Section 2.9.4 (pp. 62-63). --- 02/10/17: Lab 1 - generating random passwords. --- 03/10/17: More on generation of random permutations, 1970 draft lottery problem. Textbook: Section 2.8.1 (pp. 55-56), Section 3.7 (pp. 98-102). --- 04/10/17: Frequentist probability, Axiomatic foundations of probability. Textbook: Chapter 7, Section 7.1 (pp. 229-232). --- 09/10/17: Lab 2 - does skill win the league? --- 10/10/17: Definition of probability functions on a finite or countably infinite sample space, Equally likely events, Examples. Textbook: Section 7.1.1 (pp. 232-235). Assignment: Problems 7.1-7.5 and 7.14 (if you want them to be checked -not marked- hand your solutions in on Wednesday 18/10/17). --- 12/10/17: Probability on an uncountable sample space (dartboard example), Derivation of basic rules of probability from axioms, Independent events. Textbook: Example 7.3 (pag. 234), Section 7.2 (pp. 239-241, 7.2.1 excluded), Section 7.3 (pp. 243-244). Assignment: Probl. 7.27 and 7.28 (if you want them to be checked -not marked- hand your solutions in on Wednesday 18/10/17). --- 17/10/17: Monty Hall problem via simulation, Conditional probability, Chain rule, Bayes rule. Textbook: Chapter 6 (pp. 212-217, 6.1.3 excluded, pp. 220-222), Chapter 8 (pp. 256-258, pp. 260-261). Assignment: Problems 8.1-8.3 (if you want them to be checked -not marked- hand your solutions in on Tuesday 31/10/17). --- 18/10/17: Tutorial: solution of Exercise 7.3, 7.4, 7.5 and 7.14. Binomial coefficient. --- 23/10/17: Lab 3 - system lifetime. --- 24/10/17: Law of conditional probability (or law of total probabilities), Bayes theorem, Monty hall problem with Bayes theorem, Problems 8.30 and 8.19. Textbook: pp. 264-267. Assignment: 8.17, 8.31, 8.32 (if you want them to be checked -not marked- hand your solutions in on Tuesday 31/10/17). --- 25/10/17: Random variables (RV), Range of a RV, Probability mass function (or distribution) of a discrete RV, Expected value of a discrete RV. Textbook: Chapter 9, pp. 283-290. Assignment: 9.1, 9.2, 9.3 (if you want them to be checked -not marked- hand your solutions in on Tuesday 31/10/17) --- 31/10/17: Linearity of expectation: E[X+Y]=E[X]+E[Y] (proof for discrete R.V.s); Expectation of g(X); Counterexample disproving E[g(X)]=g(E[X]); E[aX+b]=aE[X]+b. Textbook: pp. 290-294. Assignment: 9.14. --- 01/11/17: Measures of dispersion: variance and standard deviation of a R.V.; Equivalent definitions for the variance; Independent R.V.s (general case); Textbook: pp. 294-295, p. 299. Assignment: prove that Var(aX+b)=a^2Var(X) (prove it for X discrete). --- 13/11/17: Lab 4 - Summaries of variation. --- 14/11/17: Independent R.V.s (discrete case); X and Y independent => E[XY]=E[X]E[Y] (but not viceversa); Distribution of sum of independent R.V.s (discrete case); Bernoulli distribution. Textbook: pp. 300-301, convolution formula at p. 302, 303-304. Assigment: - Toss of two dice: let X=sum of outcomes, Y=max of outcomes, and prove that X and Y are not independent. - Prove that if X and Y are independent then Var(X+Y)=Var(X)+Var(Y). --- 15/11/17: Binomial distribution; Geometric distribution. Textbook: pp. 304-305; Section 9.6.5. at p. 312. Assignment: Problem 9.32. --- 20/11/17: Geometric sum and geometric series; Poisson distribution; Poisson approximation of the binomial distribution (no proof). Textbook: pp. 111-113; Assignment: Verify that if X is Geometric(p), the sum for k that goes from 1 to infinity of P(X=k) is equal to 1. --- 21/11/17: Cumulative distribution function; Uniform distribution; Continuous random variables; Probability density function Textbook: pp. 318-322, pp. 333-334. --- 22/11/17: Density function is not unique; From cumulative distribution function to density; Interpretation of density functions. Textbook: pp. 322-326. --- 27/11/17: If X is continuous, then P(X=x)=0 for every x; Exponential distribution; Exponential r.v. as transformation of a uniform r.v.; inverse-transform method. Textbook: pp. 335-336, pp. 353-354. --- 28/11/17: Expectation for a continuous r.v. X; Expectation of g(X); Memoryless property of the exponential distribution; Connection between exponential and Poisson; Normal distribution. Textbook: pp. 326-330, pp. 336-339, pp. 343-344. --- 29/11/17: Standard normal distribution; Standardisation; Joint probability mass function of two discrete r.v.'s X and Y; Marginal distributions; Probability table. Textbook: pp. 344-345, pp. 360-362. Assignment: Problems 10.11, 10.36. --- 30/11/17: Joint density of two continuous r.v.'s X and Y; Interpretation of the joint density; Marginal densities; Convolution formula (no proof). Textbook: pp. 362-368. Assignment: Problem 11.12. --- 06/12/17: Tutorial by Fearghal Donaghy. Solution to mock exam; Covariance of two random variables; If X and Y are independent => Cov(X,Y)=0 but not viceversa. --- 11/12/17: Independent random variables (continuous case); Expected value of g(X,Y); X and Y continuous independent RVs => E[XY]=E[X]E[Y]; X and Y continuous RVs => E[X+Y]=E[X]+E[Y]; Simple linear regression. Textbook: pp. 367-374 --- 12/12/17: Least squares method: derivation of optimal regression coefficients. Geometric interpretation of least squares line. Textbook: / --- 13/12/17: Properties of residuals: null mean and zero correlation with the predictor; Decomposition of variance; coefficient of determination. Textbook: / --- 14/12/17: Normal linear regression; 95% prediction intervals for the response variable. Textbook: / Exercise on regression |