Text Box: University of New Hampshire - The Graduate School www.gradschool.unh.edu Department of Mathematics and Statistics www.math.unh.edu/ Graduate Certificate in INDUSTRIAL STATISTICS Program Requirements A Graduate Certificate in Industrial Statistics is awarded for completion of three courses as follows: Two Required Courses: Any two of: Math 837: Statistical Quality Improvement (SQI) Math 839: Applied Regression Analysis Math 840: Design of Experiments I Math 844: Design of Experiments II (DOE II) One Elective Course: chosen from remaining courses of the above list and from Math 841: Biostatistics and Life Testing Math 842: Multivariate Statistical Methods or any other approved special topics course in the area of industrial statistics • Other special topics courses are occasionally offered and may be added to the list of elective courses. • Most of these courses are currently offered live over the Internet in "Far View" mode. See www.unh.edu/farview/ for more information on the Far View program. Proposed Schedule of Offerings Fall 2003: 839, 844; Spring 2003: 840, 842, 896(Nonparametric Statistics); Summer 2004: 837 Application Please fill out an application and a residency form and submit these plus undergraduate transcripts to the Graduate School. These forms are available on-line in Acrobat Reader (pdf) format: Application Form , Residency Form For general information, see the Graduate School's website: http://www.gradschool.unh.edu/ Program Coordinator Ernst Linder, M303 Kingsbury Hall, UNH, Durham, NH 03824 Tel: 603-862-2687, Fax: 603-862-4096 email: elinder@math.unh.edu Answers to Frequently Asked Questions about Graduate Certificates • Individuals holding a Bachelor's degree are eligible to apply for admission to a graduate certificate program • Applicants may or may not already be enrolled in a graduate degree program at UNH. • Application for the certificate should be made in the first two semesters of coursework. All course work for the certificate must be completed within three years from the date of matriculation. • A maximum of one course taken prior to matriculating may be applied to the certificate. • Courses may be applied to only one certificate program but may be applied to a masters or Ph.D. program as appropriate. • Tuition for in-state students is the same as per credit rate for in-state graduate degree students. Out-of-state rate is 10% higher. • A grade of B- or higher must be earned in these courses in order for them to be counted towards the certificate. Statistics Program Faculty and their Area of Expertise Zhaozhi Fan, Ph.D. (2001) Heavy – Tailed Distributions, Longitudinal Data Marie Gaudard, Ph.D. (1977) Design of Experiments, Process Control, Quality Improvement, Data Mining Linyuan Li, Ph.D. (2002) Wavelets, Life Time Distributions, Long Memory Processes Ernst Linder, Ph.D. (1987) Environmental Statistics, Spatial & Spatial Temporal Statistics, Geostatistics, Bayesian and Computational Statistics Philip Ramsey, Ph.D. (1989) Design of Experiments, Statistical Process Control

 

Course Descriptions


Math 837: Statistical Quality Improvement

 

Introduction to scientific data collection and analysis with an emphasis on industrial applications. Topics include SPC, engineering process control, failure modes and effects analysis (FMEA), Six-Sigma concepts and methods, and confidence intervals and hypothesis testing. Use of a statistical software package is an integral part of the course; graphical data analysis will be emphasized. Prerequisites: Permission by instructor. 3 cr.

 

Math 839:  Applied Regression Analysis

 

Regression analysis explores relationships among variables by modeling a response. Simple linear regression, residual analysis and model validation, multiple linear regression, model selection, multicollinearity, nonlinear curve fitting, categorical predictors, introduction to analysis of variance and covariance. Statistical software will be used extensively. Prerequisites: Math 837 or permission 3 cr.

 

Math 840:  Design of Experiments I

 

This course emphasizes methods for solving complex problems, both in the industrial and research environments. Design of experiments, randomization and blocking, factorial designs, nested designs, fixed, random, and mixed effects models, fractional factorial designs, use of covariates, response surface methods. JMP software is used extensively. Prerequisites: Math 837 or permission. 3 cr.

 

Math 841:  Biostatistics and Life Testing

 

Exploration of models and data-analytic methods used in medical, biological, and reliability studies. Event-time data, censored data, reliability models and methods, Kaplan-Meier estimator, proportional hazards, Poisson models, loglinear models. SAS or JMP, and SPlus will be used. Prerequisites: Math 837 or permission. 3 cr.

 

Math 842:  Multivariate Statistical Methods

 

Issues dealing with multivariate response data. Random vectors and matrices, multivariate normal distribution, Hotelling's T2, multivariate analysis of variance (MANOVA), principal components, cluster analysis, factor analysis, longitudinal data and repeated measures. SAS or SPlus will be used. Prerequisites: Math 837 or permission. 3 cr.

 

Math 844:  Design of Experiments II

 

This course will focus on experimental design strategies and issues that are often encountered in practice, but that are typically not covered in an introductory course. Participants will develop a high degree of expertise and proficiency in experimental design. Industrial situations will be the focus of the course, and participants will be encouraged to bring their experimental design issues to class. Although the course will review the basics of experimental design and analysis, participants should have some prior experience with factorial experiments and their analysis. Prereq: MATH 840 or permission. 3 cr.