Course Handbook. Postgraduate Diploma in Statistics 2009/2010. School of Computer Science and Statistics

Transcription

1 Course Handbook Postgraduate Diploma in Statistics 2009/2010 School of Computer Science and Statistics Trinity College Dublin

2 September, Dear Participant, Welcome to the Postgraduate Diploma in Statistics. This booklet contains some important information on various aspects of the course - dates of lecture terms, examination regulations and course outlines. So please take some time to read it. The course will be taught jointly by a group of people. See attached list of modules for information on the lecturers and content of the modules. I can be contacted at , Eamonn.Mullins@tcd.ie throughout the year if you have any queries or problems in relation to the academic side of the course. Natasha Blanchfield is the executive officer who administers the course; her number is , Natasha.Blanchfield@scss.tcd.ie. Administrative queries should be addressed to Natasha in the first instance. Note that we will want to communicate with you by from time to time. We will use the College address that you will be assigned (many of you will have these already in the form NAME.tcd.ie). It is important, therefore, that you check your college mail regularly (and frequently!). It is important that you register at the appropriate times if you do not, you may be charged a late registration fee Unfortunately we are unable to provide parking on campus. I hope you enjoy the course. Best wishes, Eamonn Mullins Course Director Disclaimers: The information contained in this document is intended to provide a guide to those seeking admission to the programme, and to the students on the course. Trinity College Dublin reserves the right to update or change syllabi, timetables, or other aspects of the programme at any time. Changes will be notified to current students by .

3 TABLE OF CONTENTS FOR COURSE HANDBOOK. DIPLOMA IN STATISTICS ACADEMIC STRUCTURE... 4 DISTINCTIONS... 4 EXAMINATIONS... 4 DEFERRING EXAMINATIONS...5 TAUGHT MODULES... 6 INDIVIDUAL WORK AND PLAGIARISM... 6 MODULE DESCRIPTIONS... 7 STAFF COURSE DIRECTOR COURSE ADMINISTRATION EXTERNAL EXAMINER LECTURERS APPENDIX 1: TIMETABLE APPENDIX 2: TCD WEB LINKS APPENDIX 3: MAPS... 16

4 Academic Structure The Diploma consists of a Base Module and a series of elective modules. The elective modules offered may vary from year to year. To obtain the Diploma participants must pass the Base Module and two elective modules. This is normally done within one academic year, but to facilitate people who, due to work commitments, cannot do this, it is possible to take the course over two years. Note, though, that fees are charged for the second year. Students who wish to take the Diploma course over two years must apply to do so before the end of the first semester. Distinctions Each module will be graded as follows: fail (< 50%), pass (50%+), and distinction (75%+). To obtain the Diploma with distinction, participants must obtain an overall average of at least 75%. This may include one grade between 70% and 74%; students who obtain less than 70% in any module will not be awarded a distinction even if their overall average exceeds 75%. Distinctions are not awarded at supplemental examinations. Examinations Each module will be examined separately. The form of the examination may vary from module to module and may include assignments, written examinations or both. Details will given by the lecturers concerned. Students take the examinations for all modules during the annual examination period. This will be between April 25th and 21st May, Individual examination dates are set by the Examinations Office of the University and will not be available until the second semester. Examination timetables will be posted on the College website: No letters will be sent regarding this matter. Subject to the recommendation of the court of examiners, students who are unsuccessful in the annual examinations may be allowed a supplemental examination. The dates for the supplemental examination period are: 30 th August 10 th September,

5 Deferring Examinations In exceptional circumstances, permission may be granted to defer examinations to the supplemental examination period. Participants must apply to the Course Director for such permission at least one month prior to the examination date. If a student cannot sit an examination for medical reasons, medical certificates have to be submitted to the Course Administrator; the College regulations require that this be done within three days of the beginning of the period of absence from the examination. Where deferral is granted for either medical or non-medical reasons and the student fails the examination in the supplemental period, the student may be granted permission to sit a supplemental examination for the failed module(s) the following year. In such cases the student must register and pay the appropriate fee for the second year. Research students who are presenting papers at conferences, or who need to undertake fieldwork which cannot be carried out at another time, must supply a supporting letter from their research supervisor when applying for a deferral. Please note that inadequate preparation for the examinations is not a valid basis for a deferral. If a student is not adequately prepared and does not wish to attempt one or more examinations, they may apply to the Course Director to take the supplemental examination. Note that this will mean that they will not be allowed a second attempt should they fail the examination(s). Failure in the supplemental examination for the Base Module (i.e., on a second attempt) leads to exclusion from the course. Students who fail a supplemental examination in an elective module may take a different elective module (but only if one is available, which is not guaranteed) in the second year of the course. This may be done only once, i.e., at most three elective modules may be attempted. Students who have not been given permission to defer and who do not present for an examination are automatically excluded from the course. 5

6 Taught Modules The modules being run this year are shown below. Module Code Module Name Module Coordinator ST7001 Base Module Mr E Mullins ST7002 ST7003 ST7003 Introduction to Regression Design and Analysis of Experiments Time-series Analysis Sem ester 1 Prof. J Haslett 2 Dr M Stuart 2 Dr R Dahyot 2 ECTS Assessment 15 Examination 10 Examination 10 Examination 10 Examination Individual Work and Plagiarism It is important to highlight that all work submitted must be your own, and not taken literally from the internet or other sources. The regulations governing plagiarism are available in the college calendar ( Copies are held in the College Library, Enquiries Office, and all academic and administrative offices. In the case of group work, groups should establish some mechanism to ensure that no member engages in plagiarism. Note that lecturers or the course director may submit any piece of submitted work to the TurnItIn plagiarism detection tool which detects any plagiarism of web material and of any other material previously submitted to TurnItIn. (See 6

7 Academic Year Module Code ST7001 Module Title Base Module ECTS 15 Chief Examiner Mr Eamonn Mullins Teaching Staff Mr Eamonn Mullins Delivery Course Outline Classroom teaching; lectures/computer labs For those who know what the names mean, the headings below give an outline of the topics that will be discussed in the base module; I hope everyone will know what they mean by the end of the course! The base module, as the name suggests, is introductory and will lay down the foundations on which other modules will build. The fundamental statistical inferential ideas of significance tests and confidence intervals are the central topics. The various inferential methods will be unified through the concept of a statistical model, which is an abstract representation of the quantity we wish to describe. For example, we may choose to represent the weights of filled containers by a Normal distribution with a particular centre (mean) and measure of spread (standard deviation). This would allow us to introduce formal tests to determine when the process average weight changes. Of course, the value of any formal procedure will depend on how well the underlying model represents the characteristics of the practical problem. When models are fitted, good statistical practice requires the assessment of the models used; this is done mainly by use of graphical procedures. These may be simple scatterplots of two characteristics of a number of individuals (e.g., heights and weights of a sample of people) to determine whether or not the assumption of a linear relation between the two characteristics is reasonable. Alternatively, the graph might be a Normal probability plot (quantile-quantile plot) of residuals (differences between observed and predicted values) after a complex multiple regression model has been fitted to the data. Many questions can be answered by simple plots. Therefore, the apparently simple intuitive graphical tools are just as important as the apparently more sophisticated mathematical techniques. Note that complicated does not equal sophisticated when it comes to statistical practice: there are many examples of the plainly silly use of complicated statistical techniques in practically every empirical discipline. Note that as there are separate modules on both regression and design of experiments (the main methods of analysis involve analysis of variance), the coverage of these topics will be limited to the most elementary models. 7

8 Learning Outcomes Syllabus On successful completion of the Base Module students should be able to: demonstrate a systematic understanding of the fundamental inferential ideas which underpin statistical methods demonstrate a broad understanding of the role of statistical ideas and methods covering both data collection and data analysis demonstrate a competence in the use of basic statistical tools They will have a sound basis on which to develop further their statistical skills. Specific topics addressed in this module include: Data summaries and graphs Statistical models Sampling distributions: confidence intervals and tests Comparative experiments: t-tests, confidence intervals, design issues Counted data: confidence intervals and tests for proportions, design issues Cross-classified frequency data: chi-square tests Introduction to Regression Analysis Introduction to Analysis of Variance Statistical computing laboratory Assessment Bibliography One 3-hour examination & optional coursework The following book is a suitable general reference for the base module. I will give extensive handouts during the semester, so that the course will not be based on Moore and McCabe; it will, however, provide a useful second view on what statistics is about. D.S. Moore and G. P. McCabe, Introduction to the practice of statistics, Freeman, 5th edition, 2006 My own book was written for analytical chemists, but it would be suitable reading for most natural scientists and engineers. Moore and McCabe would be more suitable for social scientists. E. Mullins, Statistics for the quality control chemistry laboratory, Royal Society of Chemistry, Those with medical interests will find the following a useful reference book: D. G. Altman, Practical statistics for medical research, Chapman and Hall, Those interested in business and industry will find lots of interesting examples in: M. Stuart, "Introduction to Statistical Analysis for Business and Industry, a problem solving approach", Hodder Arnold Publishers, Website 8

9 Academic Year Module Code Module Title ST7002 Introduction to Regression Prerequisites Diploma Base Module ECTS 10 Chief Examiner Teaching Staff Delivery Prof John Haslett Prof John Haslett Classroom teaching; lectures/computer laboratory Aims Learning Outcomes Syllabus An introduction to multiple linear regression The most widely used tool in statistics is that of regression. It is therefore one the most misused. When students have successfully completed this module they should: Understand the concepts involved in multiple linear regression Understand how to use MINITAB software for regression Understand the pitfalls in analysis Understand how and why it can be used as the basis for very many aspects of statistical analysis Understand its limitations Specific topics addressed in this module include: Review of simple linear regression model: assumptions, model fitting, estimation of coefficients and their standard errors The multiple linear regression model and its analysis including: o Confidence intervals and statistical significance tests on model parameters o Issues in the interpretation of the multiple parameters o Prediction intervals o Analysis of variance in regression: F-tests, r-squared Model validation: residuals, residual plots, normal plots, diagnostics Modern extensions Assessment One 2-hour examination Bibliography L.C. Hamilton Regression with Graphics, Duxbury Press, 1992 E. Mullins, Statistics for the quality control chemistry laboratory, Royal Society of Chemistry, M. Stuart, "Introduction to Statistical Analysis for Business and Industry, a problem solving approach", Hodder Arnold Publishers, Website 9

10 Academic Year Module Code ST7003 Module Title Design and Analysis of Experiments ECTS 10 Chief Examiner Michael Stuart Teaching Staff Michael Stuart Delivery Course Outline Classroom teaching; lectures /computer laboratories This module is concerned with the design of data collection exercises for the assessment of the effects of changes in factors associated with a process and the analysis of the data subsequently produced. In order to assure that the experimental changes caused the observed effects, strict conditions of control of the process must be adhered to. Specifically, the conditions under which the experimentation is conducted must be as homogeneous as possible with regard to all extraneous factors that might affect the process, other than the experimental factors that are deliberately varied. The simplest experiments involve comparison of process results when a single factor is varied over two possible conditions. When more than two factors are involved, issues regarding the most efficient choice of combinations of factor conditions and ability to detect interactions between factors become important. With many factors and many possible experimental conditions for each factor, the scale of a comprehensive experimental design becomes impractical and suitable strategies for choosing informative subsets of the full design are needed. The analysis of data resulting from well designed experiments is often very simple and graphical analysis can be very effective. Standard statistical significance tests may be used to assure that apparent effects are real and not due simply to chance process variation. In cases with more complicated experimental structure, a more advanced technique of statistical inference, Analysis of Variance, may be used. Confidence intervals are used in estimating the magnitude of effects. Minitab is well equipped to assist both with design set up and with analysis of subsequent data, both graphical and formal. There will be two laboratory sessions involving the use of Minitab. Case studies and illustrations from a range of substantive areas will be discussed 10

11 Learning Outcomes On successful completion of this module, students should be able to compare and contrast observational and experimental studies, describe and explain the roles of control, blocking, randomisation and replication in experimentation, explain the advantages of statistical designs for multifactor experiments, describe and explain the genesis of a range of basic experimental design structures, implement and interpret the analysis of variance for a range of basic experimental designs, describe the models underlying the analysis of variance for a range of basic experimental designs, produce and interpret graphs for data summary and model diagnostics, provide outline descriptions of more elaborate designs and data analyses, describe and discuss strategic issues involved in the design and implementation of experiments. Syllabus Assessment Specific topics addressed in this module include: Experimental and observational studies o control of the study environment o cause and effect Basic design principles for experiments o Control o Randomisation o Blocking (pairing) o Replication Standard designs o Randomised blocks o Two-level factors o Fractional factorial designs o Split unit experiments o Repeated measures experiments Analysis of experimental data o Exploratory data analysis o Parameter estimation and significance testing o Analysis of variance o Statistical models, fixed and random effects o Model validation, diagnostics o Software laboratories Review topics o Response surface designs o Analysis of Covariance o Robust designs o Non-Normal errors Strategies for Experimenting o Consultation o Planning o Resources o Ethical issues o Implementation of design o Application of results One 2-hour examinations 11

12 Reference Mullins, E., Statistics for the Quality Control Chemistry Laboratory, Royal Society of Chemistry, 2003, particularly Chapters 4-5, 7-8. Detailed coverage of much of the module, in a specific context. Montgomery, D.C., Design and analysis of experiments, 6th ed., Wiley, A general comprehensive text, covers much more than this module, including statistical theory. Not always authoritative. Box, G.E.P., Hunter, J.S. and Hunter, W.G., Statistics for Experimenters, 2nd. ed., Wiley, Includes many gems of wisdom from these masters of the genre, though not a course text. Daniel, C., Applications of Statistics to Industrial Experimentation, Wiley, Includes many gems of wisdom from this master of the genre, using methodology appropriate for an industrial setting. Robinson, G.K., Practical Strategies for Experimenting, Wiley, A comprehensive review of the non-statistical aspects of planning and conducting experiments and interpreting and using their results. Website 12

13 Module Code Module Title Pre-requisites ST7005 Time Series Analysis Basic Statistics and Mathematics ECTS 10 Chief Examiner Dr. Rozenn Dahyot Teaching Staff Dr. Rozenn Dahyot Delivery Aims Learning Outcomes Classroom teaching; lectures /computer laboratories Several methods of forecasting will be examined, including exponential smoothing and its Holt-Winters extension, autoregression, moving average, and further regression based methods that take into account seasonal trends of lagged variables. The module will be practical, and will involve every student in extensive analysis of case study material for a variety of time series data. When students have successfully completed this module they should be able to: Define and describe the different patterns that can be found in times series and propose the methods that can be used for their analysis. Program, analyse and select the best model for forecasting. Interpret output of data analysis performed by a computer statistics package. Syllabus Introduction to time series Regression Autoregressive Models Data Transformations Modelling Seasonality Decomposition Exponential Smoothing RMSE and MAPE performance measures Holt-Winters Models for Seasonality Autocorrelation Brief Introduction to ARIMA Assessment One 2-hour examination Bibliography Forecasting - Methods and Applications, S. Makridakis, S. C. Wheelwright and R. J. Hyndman, Wiley Website 13

14 Staff Course Director Mr Eamonn Mullins Tel: Course Administration Natasha Blanchfield Tel: Main Office of the School of Computer Science and Statistics Tel: External Examiner Dr David Williams, University College Dublin. Lecturers Mr E Mullins (eamonn.mullins@tcd.ie ) Tel: Prof. J Haslett (john.haslett@tcd.ie ) Tel: Dr M Stuart (michael.stuart@tcd.ie ) Dr R Dahyot (dahyotr@tcd.ie ) Tel:

15 Appendix 1: Timetable Codes, Course, Lecturer: Postgraduate Diploma in StatisticsLecture Timetable Venues: Term / Day and Time Monday Tuesday Wednesday Thursday Friday Michaelmas Term ST7001 LB04 / EEPC 1/2/ ST7001 LB04 / EEPC 1/2/ Hilary Term Wks 1-6 ST7002 LB04 / EEPC 1/2/ Hilary Term Wks 1-6 ST7002 LB04 / EEPC 1/2/3 Hilary Term Hilary Term Wks 8-12 ST7005 LB04 / EEPC 1/2/ Hilary Terms Wks 8-12 ST7003 LB04 / EEPC 1/2/ Hilary Term Wks 8-12 ST7005 LB01 / EEPC 1/2/ Hilary Term Wks 8-12 ST7003 LB04 / EEPC 1/2/3 ST7001: Base Module: Mr E Mullins LB01/4: Lloyd Institute, Lower Ground Floor, Lecture Theatre 01/ 04 ST7002: Introduction to Regression: Professor J Haslett EEPC 1/2/3: Panoz Institute, PC Lab 1/2/3 ST7003: Design and Analysis of Experiments: Dr M Stuart ST7005: Time-series Analysis: Dr R Dahyot (MT) Michaelmas Term: 28 September December 2009 (Reading Week 9-13 November 2009) (HT) Hilary Term: 18 January April 2010 (Reading 15

16 Appendix 2: TCD Web links There are many useful sites in TCD. Here are a number of them. If you find any other TCD links that you think would be useful for the class please the Course Co-ordinator. Site TCD Website Library Information System Services Graduate Studies Student Counselling Computer Science and Statistics Address Appendix 3: Maps Maps can be found online at Use the a-z search to find specific buildings. 16

17 17