Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
San Jose State University Department of Mathematics Math 164 Mathematical Statistics Catalog Description Sampling distributions, interval estimation, confidence intervals, order statistics, sufficient statistics, the Rao-Blackwell Theorem, completeness, uniqueness, point estimation, maximum likelihood, Bayes’ methods, testing hypotheses. Prerequisite Math. 163 (Probability Theory) with a grade of C- or better, or instructor consent. Textbook Larsen & Marx, (1986). An Introduction to Mathematical Statistics and Its Applications, 2nd edition, Prentice Hall. Alternative textbook Wackerly, Mendenhall & Scheaffer, Mathematical Statistics with Applications, 6th ed., Duxbury Press. References none Technology Requirement A scientific calculator which has an exponential key (yx) and a factorial key (x!) is needed for some of the homework assignments as well as for the exams. A graphing calculator, such as the TI-82, or TI-85, is useful but is not required. Course Content Brief review of probability theory. Sampling distributions, order statistics, estimation of parameters, properties of estimators: unbiasedness, efficiency, consistency, minimum variance; maximum likelihood, confidence intervals, hypothesis testing, Central Limit Theorem, and goodness-of-fit tests. Student Outcomes Students should be able to: 1. Determine whether an estimator for a parameter is biased or unbiased (for ), and what this means. 2. Calculate the relative efficiency of two estimators. 3. Determine whether an estimator (for ), W, is efficient by calculating the Cramer-Rao lower bound and Var(W) and comparing the two values. 4. Given a density function with an unknown parameter , find an efficient unbiased estimator for by using the Cramer-Rao theorem. 5. Determine whether an estimator is consistent, asymptotically unbiased, and/or squared-error consistent (for ). 6. Determine whether an estimator is sufficient (for ) by using the Fisher-Neyman criterion or by using the Factorization Theorem. 7. Calculate the likelihood function and use it to find the maximum-likelihood estimator(s) (for one or more parameters). 8. Find the method of moments estimator(s) (for one or more parameters). 9. Find confidence intervals for parameters (p, , 2, , etc.) of various distributions. 10. Do one- and two-sample tests of hypotheses involving parameters of various distributions and calculate p-values, type I and type II errors. Specifically, be able to do tests involving the standard normal distribution, the Chi-square distribution, t-tests, and F-tests. 11. Find the critical region for a test given the probability of a type I error and vice-versa. 12. Calculate the generalized likelihood ratio and find the form of the corresponding generalized likelihood ratio test. 13. Use the Central Limit Theorem to calculate probabilities involving the sample mean (or sample sum) of a sufficiently large sample. 14. Do a goodness-of-fit test for a distribution in the two cases: where all the parameters are known and where they are unknown. 15. Do a chi-square test to determine whether two traits are independent based on data given in a contingency table. Topics and suggested course schedule All sections refer to the text by Larsen and Marx. Topic Ch. 1-4: Probability Review: discrete and continuous random variables, pdfs, cdfs, expectation, variance, standard deviation and their properties, moments, central moments. Ch. 1-4: Probability Review: the major discrete distributions: Bernoulli, binomial, geometric, negative binomial, hypergeometric, Poisson; and Poisson as a limit of Bin(n,p) Ch. 1-4: Probability Review: the major continuous-type distributions: uniform, exponential, gamma, normal; and the relationship of the exponential and gamma distributions to Poisson processes § 3.6: Order Statistics § 5.1-5.4: Estimation of Parameters. Number of 75-minute lectures 1 1.5 1 1 1.5 § 5.5: Efficiency. § 5.6: Minimum Variance Estimators: The Cramer-Rao Lower Bound. § 5.7a: Consistency. Include a review of Chebyshev’s inequality. § 5.7b: Sufficiency. Include a review of conditional density and conditional expectation (§ 3.7) if covered. Optional (inclusion requires cutting the time allotted to some other topic): § 5.8: Finding Estimators: (I) The Method of Maximum Likelihood. § 5.8: Finding Estimators: (II) The Method of Moments. § 5.9: Interval Estimation (Confidence Intervals) § 5.10: Confidence Intervals for the binomial parameter p . § 6.1-6.2: Hypothesis Testing: The Decision Rule. § 6.3: Type I and Type II Errors. § 6.4: The Generalized Likelihood Ratio Test. § 7.1-7.2: Point Estimates for and 2 for the normal distribution Review: § 3.12: Moment generating functions. § 7.3: Linear Combinations of Normal Random Variables. § 7.4: The Central Limit Theorem. § 7.5: The Chi-square Distribution; Inferences about 2 § 7.6: The F Distribution and t Distribution. § 7.7: The One-Sample t Test. § 8.1-8.2: Testing H0: X = Y: The Two-Sample t Test. .5 1 1 (2) Optional (inclusion requires cutting the time allotted to some other topic): § 8.3: Testing H0: X2 = Y2: The F Test. § 8.4: Binomial Data: Testing H0: pX = pY. (1) § 8.5: Confidence Intervals for the Two-Sample Problem. Review: § 9.1-9.2: The Multinomial Distribution; § 9.3: Goodness-of-Fit Tests. § 9.4: Goodness-of-Fit Tests: Parameters Unknown. § 9.5: Contingency Tables. Exams TOTAL 1 .5 1.5 1 1 2 30 Prepared by: Amy Rocha Chair, Probability and Statistics Committee Department of Math, SJSU November 2003 1 1 .5 .5 1 1 1 1 .5 .5 1.5 1 1 1 1 1