Modules Descriptions

 (Updated on 03 August 2018) The modules listed below are offered for the Graduate Programme in Statistics. Not all modules are available in any one semester. ST5201 and ST5202 are compulsory modules specifically meant for the part-time MSc Programme (MSc by Coursework). Departmental approval is required for other students to read these modules.   ST5198 Graduate Seminar Module Modular Credits: 4 Workload: 3-1-0-3-3 Pre-requisite: Departmental approval  Preclusion: Nil The objective of this module is to introduce students to recent advances in statistical research. Students may be asked to conduct some of the seminars so as to foster more interaction between the instructor(s) and students.   ST5199 Coursework Track II ProjectModular Credits: 16 Workload: 0-2-0-0-13Pre-requisite: Nil Preclusion(s): Nil Cross-listing(s): Nil The objectives of the course are to develop the basic skills for independent scientific research, and to promote an appreciation of the application of problem solving strategies in science. On completion of the course, students will be able to demonstrate an appreciation of the current state of knowledge in a particular field of research, to master of the basic techniques required for the study of a research question, and to communicate scientific information clearly and concisely in written and spoken English.    ST5201 Statistical Foundations of Data Science Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approval Preclusion(s): Nil The module introduces basic theories and methods in Statistics that are relevant for understanding data science. Exploratory data analysis including heat map and concentration map. Random variables. Joint distributions. Expected values. Limit theorems. Estimation of parameters including maximum likelihood estimation, Bayesian approach to parameter estimation. Testing hypotheses and confidence intervals, bootstrap method of finding confidence interval, generalized likelihood ratio statistics.  Summarizing data: measures of location and dispersion, estimating variability using Bootstrap method, empirical cumulative distribution function, survival function, kernel probability density estimate. Basic ideas of predictive analytics using multiple linear and logistic regressions.      ST5202 Applied Regression Analysis Modular Credits: 4 Workload: 3-1-0-3-3 Pre-requisite: Departmental approval  Preclusion(s): ST5318 Multiple regression, model diagnostics, remedial measures, variable selection techniques,non-leastsquares estimation, nonlinear models, one and two factor analysis of variance, analysis of covariance,linear model as special case of generalized linear model. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisities.   ST5203 Design of Experiments for Product Design and Process Improvements Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approvalPreclusion(s): ST5318 The module introduces designed experiment as a tool for process improvements and designing products that are robust to environmental variability. Inferences about the effect of factors on a product or process can be drawn using designed experiment. Topics include analysis of variance of fixed-effect models, randomized block design, factorial designs, fractional factorial designs, blocking and confounding, response surface methodology, random effects models, nested and split-plot designs. Predictive analytics using designed experiments. This module is targeted at students who are interested in designing robust products and process improvements, and are able to meet the pre-requisites.    ST5204 Experimental Design 2Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approvalPreclusion(s): Nil General linear model approach to the analysis of experimental designs, means model and effects model, unbalanced designs and missing values, split-plot, strip-plot, repeated measures, nested and crossover designs, response surface methodology. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.   ST5205 Probability Theory IIModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approvalPreclusion: MA5260  Conditional expectation given a sigma-algebra, Martingale, stopping time, Martingale convergence theorems, Doob's Stopping Theorem and applications, Brownian motion, construction and sample path properties. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.   ST5206 Generalized Linear Models Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST4233 or Departmental approval Preclusion(s): Nil Model fitting and selection, models for continuous and discrete data, models for polytomous data, log-linear models, conditional and quasi-likelihoods, diagnostics. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.      ST5207 Nonparametric RegressionModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST3131 or Departmental approval Preclusion(s): Nil Various smoothing methods, including kernel, spline, nearest neighbour, orthogonal series and penalised likelihood. This module is targeted at students who are interested in Statistics and are able to meet the prerequisites.   ST5208 Analytics for Quality Control and Productivity ImprovementsModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST3235 or Departmental approval Preclusion(s): Nil This module covers modern procedures for quality monitoring and productivity improvements. Quality control charting procedures: Shewhart, cumulative sum, straight moving average and exponentially weighted moving average charts. Run length distributions.  Optimal charting procedures. Risk-adjusted charting procedures. Multivariate charting procedures. Process capability analysis. Design of experiments and process optimization. Predictive analytics of processes. Acceptance sampling procedures.  This module is targeted at students who are interested in quality control and productivity improvements, and are able to meet the pre-requisites.    ST5209 Analysis of Time Series DataModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST3233 or Departmental approval Preclusion(s): Nil Stationary processes, ARIMA processes, forecasting, parameter estimation, spectral analysis, non-stationary and seasonal models. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.     ST5210 Multivariate Data AnalysisModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST3240 or Departmental approval Preclusion(s): Nil Dimension reduction, cluster analysis, classification, multivariate, dependencies and multivariate statistical model assessment with emphasis on non-normal theory, computer intensive data-dependent methods. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.      ST5211 Sampling From Finite PopulationsModular Credits: 4Workload: 3-1-0-3-3Pre-requisite:ST2132 or Departmental approval Preclusion(s): Nil Survey data, basic sampling, stratified sampling, cluster sampling, double sampling, systematic sampling, non-response and missing values, multiple imputations, bootstrap of sampling error. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.      ST5212 Survival AnalysisModular Credits: 4Workload: 3-1-0-3-3Pre-requisite:ST2132 or Departmental approval Preclusion(s): Nil Censoring, probability models for survival times, graphical procedures, Inference procedures. Parametric and nonparametric models, Cox proportional hazards model, regression models for grouped data, Bayesian predictive distributions. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.      ST5213 Categorical Data Analysis II Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite:ST3131 or Departmental approval Preclusion(s): Nil Categorical response data and contingency tables, loglinear models, building and applying loglinear models, loglinear and logit models for ordinal variables, multinomial response models. This module is targeted at students who are interested in Statistics and are able to meet the prerequisites.     ST5214 Advanced Probability Theory  Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST2131 or Departmental approval (Compulsory to MSc by Research)Preclusion: MA5259   Probability measures and their distribution functions. Random variable: properties of mathematical expectation, independence, conditional probability and expectation. Convergence concepts: various modes of convergence of sequence of random variables; almost sure convergence, Borel-Cantelli Lemma, uniform integrability, convergence of moments. Weak and strong law of large numbers. Convergence in distribution, characteristic function: general properties, convolution, uniqueness and inversion, Lindeberg conditions and central limit theorem. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.      ST5215 Advanced Statistical Theory Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite:  ST2131 and ST2132 or Departmental approval (Compulsory to MSc by Research) Preclusion(s): Nil This module provides the students with theoretical foundations in statistics and the theory of statistical point estimation. It consists of four parts. 1. Selected Topics in Probability: convergence modes and stochastic orders; convergence in distribution; convergence of transformations; law of large numbers; central limit theorem; Edgeworth and Cornish-Fish expansions. 2. Fundamentals of Statistics: Population, sample and models; Statistics, sufficiency and completeness; Elements of statistical decision theory; Statistical inference; criteria for point estimation, Neyman-Person framework of hypothesis testing, confidence sets; Asymptotic criteria of inference. 3. Unbiased Estimation: Uniform minimum variance unbiased estimators (UMVUE); U-statistics; Least square estimators (LES) in linear models; Asymptotically unbiased estimators. 4. Estimation in Parametric Models: Bayes estimators, maximum likelihood estimators and the properties of estimators such as invariance, minimaxity, admissibility and asymptotic efficiency.   ST5217 Statistical Methods for Genetic Analysis Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: LSM1102 and ST2132 and ST3236, or Departmental approvalPreclusion(s): Nil This is a level 5000 course on genetic data analysis focusing on human and population genetics. The emphasis will be in understanding the role of statistics and data analysis in modern genetics research and its applications. Numerical and computational methods will be discussed with applications to real data sets. To equip the students with the tools to conduct genetic data analysis. Topics include introduction to genetics,gene mapping,sequence data, population genetics and coalescent theory,phylogeny reconstruction,pedigree analysis,genetic epidemiology,role of genetic factors in human diseases,familial aggregation,segregation and linkage analysis, analysis of complex and quantitative traits. This module is targeted at students who are interested in Statistics and are able to meet the pre-requisites.    ST5218 Advanced Statistical Methods in Finance  Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST4245 or Departmental approval Preclusion(s): Nil The objective of the module is to familiarise the students with selected advanced methods in quantitative finance. The major topics to be covered are: (1) Realised volatility and high frequency data, (2) Risk management under heavy-tailed distributional, (3) Assumptions, (4) Independent component analysis and its applications and (5) local parametric estimation of volatility.   ST5219 Bayesian Hierarchical Modelling Modular Credits: 4Workload: 3-1-0-0-6Pre-requisite: Departmental approval Preclusion(s): Nil The objective of the module is to familiarise the students with advanced strategies for Bayesian modelling. The major topics to be covered are: (1) Bayesian treatment of non-hierarchical regression models, (2) Stimulation based computation for inference and posterior predictive model checking, (3) Multilevel structures in Bayesian regression, (4) Multilevel models and analysis of variance, (5) Prior sensitivity analysis and (6) Advanced modelling and computation using statistical software packages.   ST5220 Statistical Consulting Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite(s): ST3131 Regression Analysis and ST3232 Experimental Design or Department ApprovalPreclusion: Nil The goal of this module is to develop the skills needed by a statistical consultant. Emphasized topics include data analysis, problem solving, report writing, oral communication with clients, issues in planning experiments and collecting data, and practical aspects of consulting management.    ST5221 Probability and Stochastic Processes Modular Credits: 4Workload: 3-1-0-0-6Pre-requisite: ST2131 or its equivalent Preclusion: ST5214, MA5259Cross-listing(s): Nil This module aims to provide graduate students in the PhD Biostatistics program a solid background and a good understanding of basic results and methods in probability theory and stochastic models. These skills are relevant for them to take advanced modules in biostatistics, and to apply state-of-the-art Biostatistics research methodologies.   ST5222 Advanced Topics in Applied Statistics  Modular Credits: 4 Workload: 3-1-0-3-3 Pre-requisite: Departmenal approval  Preclusion: Nil Cross-listing(s): Nil Topics requiring a high level of statistical computing and some optimization can be covered here, for example, discriminant analysis, machine learning, high-dimensionality and false discovery rates, stochastic search, MCMC, Monte Carlo integration, kernel smoothing and EM optimization methods.   ST5223 Statistical Models: Theory/Applications  Modular Credits: 4 Workload: 3-1-0-3-3 Pre-requisite: Departmenal approval  Preclusion: Nil Cross-listing(s): Nil Univariate and multivariate regression, graphical displays, normal equations, Gramm-Schmidt orthogonalization and singular value decomposition, model selection and prediction, collinearity and variable selection, diagnostics: residuals, influence, symptoms and remedies, ANOVA, fixed and random effects, nonlinear models including logistic regression, loglinear models and generalized linear models, computations with datasets using statistical computer package.   ST5224 Advanced Statistical Theory II Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST5215 or Departmenal approval Preclusion: NilCross-listing(s): Nil Confidence intervals, P-values, classical (Neyman-Pearson) tests, UMP tests, Likelihood ratio test, Power, Wald's test, Rao's Score test, Application of likelihood ratio tests to regression. Additional topics that can be covered in this module includes resampling methods, Bayes procedures, robustness, time series, empirical and point processes, optimal experimental design, parametric, semi-parametric and non-parametric modelling, survival analysis and sequential analysis.   ST5225 Statistical Analysis of Networks Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST2131Co-requisite: ST5201Preclusion: NilCross-listing(s): Nil Network data has become increasingly important in both academia and industry. Many interesting questions can be understood and analysed through networks. Applications are found in areas wuch as sociology (Facebook and Twitter networks), computer science (World Wide Web), and biology (gene and protein interaction networks). With the availability of large network data sets, be it in corporate, governmental or scientific contexts, comes the necessity to work with such data in an appropriate matter. This course gives a practical introduction to the theory of network analysis; topics include statistical network models, descriptive and inferential network analysis, network visualisation.   ST5226 Spatial Statistics Modular Credits: 4 Workload: 3-1-0-3-3 Pre-requisite: ST5201 Co-requisite: ST5201 Preclusion: Nil Cross-listing(s): Nil At present, almost all data that is collected is stamped with a location. This spatial information can help us in our understanding of the patterns in the data. The course is designed to introduce students to methods for handling and analysing such data. Topics covered include basic concepts of spatial data, prediction (kriging) for stationary data, and modeling the three main types of spatial data - geostatisical, areal and point pattern. R will be extensively used to demonstrate and implement the techniques.   ST5227 Applied Data MiningModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: ST5201 and ST5202Co-requisite: ST5201Preclusion: NilCross-listing(s): Nil Data mining is an interdisciplinary science to discover useful structure, to extract information from large data sets, and to make predictions. This module will focus on the most recent but well accepted methods, especially those in investigating big and complicated data, including Lasso regression, nonparametric smoothing, Neural Networks and machine learning. This module is targeted at students who are interested in handling large and complicated data sets and are able to meet the pre-requisites.   ST5241 Topics I Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approval Preclusion(s): Nil This module consists of selected topics which may vary from year to year depending on the interests and availability of staff.      ST5242 Topics IIModular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approval Preclusion(s): Nil This module consists of selected topics which may vary from year to year depending on the interests and availability of staff.     ST5243 Topics III Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approval  Preclusion(s): Nil This module consists of selected topics which may vary from year to year depending on the interests and availability of staff.     ST5244 Topics in Data Science and Analytics Modular Credits: 4Workload: 3-1-0-3-3Pre-requisite: Departmental approval  Preclusion(s): Nil This module consists of selected topics which may vary from year to year depending on the interests and availability of staff.       ST5318 Statistical Methods for Health SciencePre-requisite: MDG5108 or ST2238 or PR2103 or equivalent, or Departmental ApprovalPreclusion(s): ST5202, ST5203The module aims to equip the students with a repertoire of statistical methods that they will find useful in their research. Major topics include multiple regression, experimental designs, categorical data analysis, analysis of repeated measures, logistic regression, discriminant and classification analysis, unified approach for correlated data analysis, survival data analysis.