Graduate

Classification : Requisite   Class Number  :  326.513   Credit : 3   Grade : Graduate School

This course deals with probability measurement theory, basics of integration, random variables, independence, various modes of convergence of random variables, random series, law of large numbers, convergence in distribution, characteristic functions and central limit theorems.

Classification : Elective   Class Number  :  326.516   Credit : 3   Grade : Graduate School

This course primarily focuses on conditional probability of random variables and martingale theories, covering convergence theorems, inequalities, decompositions, optional sampling theorems, central limit theorems, as well as topics like uniform integrability and infinite divisible distributions.

Classification : Elective   Class Number  :  *326.517A   Credit : 3   Grade : Graduate School
This course aims to develop the quality skills of a statistician. Students will improve their abilities to solve statistical problems in various fields of social science as well as those of natural science. Under the professor’s supervision, students are required to present outcomes on real statistical problems and have group discussion.

Classification : Requisite   Class Number  :  326.519A   Credit : 3   Grade : Graduate School

This course covers statistical inference based on measure theory. After covering the basic concepts of sufficiency, exponential family, and various modes of convergence, the course moves on to the theory of estimation and testing. Estimation methods include the method of moment, maximum likelihood estimator, Bayes estimation, M-estimator, and Z-estimator. We derive the asymptotic distribution of these estimators and prove an efficiency theorem of the maximum likelihood estimator. Testing methods include the maximum likelihood ratio test and its asymptotic approximation, the Rao test and the Wald test as well as the Bayesian test.

Classification : Requisite   Class Number  :  326.520A   Credit : 3   Grade : Graduate School

This course explores the role of linear models as a statistical tool for modeling data. Theoretical aspects of such models are explored, but the emphasis is on strategies and methodology for model selection, estimation, inference and checking. Models covered include simple and multiple regression, and one-way and two-way analysis of variance for factorial experiments. Inference will be based largely on the least-squares criterion, exploiting the Gauss-Markov theorem, but connections will also be made with likelihood-based approaches. The use of R for modeling data via linear models will be integral to the course.

Classification : Elective   Class Number  :  326.521   Credit : 3   Grade : Graduate School

This course aims to give you an overview of the development of algorithms and statistical inference, which is a more generalized concept of exploratory and confirmatory data analysis in the era of big data. First, you will learn about the three major schools in statistics: Bayesian, Frequentist, and Fisherian, and then you will learn the modern statistical methodologies such as cross-validation and model selection, resampling methods, shrinkage estimation, empirical Bayes, resampling, survival analysis and EM algorithms, MCMC, and multiple comparisons.

Classification : Elective   Class Number  :  326.522   Credit : 3   Grade : Graduate School

This course introduces basic notions and properties regarding the weak convergence of the sequences of random elements taking values in Hilbert, in C[0,1] and in D[0,1] spaces. It also covers some important topics in empirical process theory such as concentration inequalities, uniform convergence and asymptotic equicontinuity. In addition, it introduces some useful techniques of deriving lower bounds on minimax risks in statistical estimation.

Classification : Elective   Class Number  :  *326.621A   Credit : 3   Grade : Graduate School
This course consists of series of seminars on emerging statistical theory.

Classification : Elective   Class Number  :  326.626A   Credit : 3   Grade : Graduate School

This course introduces basic methodology and theory for estimating nonparametric models. In particular, it covers kernel-based methods of estimating probability density, mean regression and quantile regression functions. In regression estimation, the coverage includes various kernel smoothing techniques based on the Nadaraya-Watson idea, local polynomial approximation and quasi-likelihood. The course also introduces three main approaches to using spline functions, namely regression spline, penalized spline and smoothing spline methods. In addition, it deals with estimating nonparametric structural regression models such as additive models and partially linear models.

Classification : Elective   Class Number  :  *326.631A   Credit : 3   Grade : Graduate School
This course consists of series of seminars on emerging applications of statistics.

Classification : Elective   Class Number  :  M1399.001300   Credit : 3   Grade : Graduate School

In this course, we study the theory of Bayesian statistics. In particular, we study constructions of noninformative priors, non-parametric Bayesian statistics, Bayesian asymptotics, and the theory for Bayesian computation.

Classification : Elective   Class Number  :  326.636   Credit : 3   Grade : Graduate School
This course covers various advanced techniques for analyzing censored and truncated survival data. The topics include Kaplan-Meier estimator, non-parametric maximum likelihood method, empirical likelihood, counting process techniques, martingales and stochastic integration, estimation of hazard function, log-rank and Gehan test, Cox’s proportional hazard model and partial likelihood.

Classification : Elective   Class Number  :  326.637   Credit : 3   Grade : Graduate School

This course deals with how to efficiently apply various advanced analysis methodologies related to data mining and big data analysis through various projects and examples. Moreover, there are team projects to learn how to implement the methods learned into data mining and big data analysis. We discuss business trends related to data mining and big data, and examine the relationship between business trends and analysis methodologies.

Classification : Elective   Class Number  :  326.638   Credit : 3   Grade : Graduate School

This course will cover statistical methods used to analyze a variety of data in genomics. The course will include a simple overview of genomic data and terminology and will proceed with a review of numerical techniques frequently employed in genomic studies. The course will focus on the statistical methods to cover topics relating to gene expression data analysis and genetic epidemiology such as linkage analysis and tests of association.
Classification : Elective   Class Number  :  326.723A   Credit : 3   Grade : Graduate School
This course is the introductory course to regression analysis for master-level graduate students. This course deals with basic matrix algebra and statistical theory, basic regression analysis, inference for simple regression analysis, miscellaneous topics for regression analysis, basic multiple regression analysis, estimation and hypothesis testing, polynomial regression, generalized regression analysis, use of dummy variables, application of analysis of variance, response surface analysis, analysis of mixture experiments, selection of variables, regression diagnostics, biased estimation, nonlinear regression and so on. The prerequisite courses are Statistics and Lab. for basic statistics and linear algebra for matrix theory.
Classification : Elective   Class Number  :  *326.739A   Credit : 3   Grade : Graduate School
This course consists of series of seminars on emerging statistical theory and application
Classification : Elective   Class Number  :  326.747   Credit : 3   Grade : Graduate School
This course provides an introduction for using statistical methods to analyze categorical data. Since categorical data can usually be arranged in a contingency table, this course focuses on using statistical methods to analyze contingency tables. The main topics in this course are contingency table analysis, log linear models, and logistic models.

Classification : Elective   Class Number  :  326.748A   Credit : 3   Grade : Graduate School

This course introduces the use of statistical methods to analyze repeated measurement data in experiment data in experimental conditions or at multiple times with one subject. It covers how to use classical multivariate models on multivariate normal distribution and mixed models to analyze continuous repeated measurement data. The course also examines how models based on weighted least squares estimation, random effect models and generalized estimating equations(GEE) are used to analyze repeated measurement data of discrete type.
Classification : Elective   Class Number  :  *326.750A   Credit : 3   Grade : Graduate School
In this course, students study recently published papers related to density, regression and frontier function estimation.
Classification : Elective   Class Number  :  M1399.000200   Credit : 3   Grade : Graduate School
Statistical computing becomes ever more important with the advent of Big Data or large scale high-dimensional data. In this course, we study the recent statistical computing techniques for large scale high-dimensional data including statistical computing using GPU and parallel computing.
Classification : Elective   Class Number  :  M1399.000300   Credit : 3   Grade : Graduate School
This course focuses on statistical methods in Spatial Statistics. The goal of the course is to learn the analysis methods for spatial and spatio-temporal data and their theoretical backgrounds and apply such methods. The contents of the course includes but not limited to hypothesis test of spatial dependence, spatial dependence models and estimation, spatial regression and kriging, analysis of areal data, disease mapping, and spatial point process models.

Classification : Elective   Class Number  :  M1399.000400   Credit : 3   Grade : Graduate School

This course builds on Advanced Methods in Data Mining (326.637) focusing on learning deep compositional functions. The goal is to study deep learning methodologies and identify related statistical issues. The contents include pre-deep learning methods such as feature extraction and discrimination; components of well-established machine learning tools (support vector machine, reproducing kernel Hilbert space, model complexity, LASSO, ensemble); neural network; multi-layer-perceptron; backpropagation; convolutional neural network; optimization and regularization; visualization; Python and deep learning frameworks; recurrent neural network; variational inference; generative adversarial network; segmentation; detection; and natural language processing.

Classification : Elective   Class Number  :  M1399.000500   Credit : 3   Grade : Graduate School

Statistical machine learning has been used popularly for data science and artificial intelligence. In this course, various methodologies of statistical machine learning are introduced and their theoretical backgrounds are explained. Supervised learning is mainly considered, and decision theory, high dimensional linear model, nonparametric function estimation, decision tree and ensemble, support vector machine and deep neural networks are included. The empirical risk minimization principle which is a principle applicable to most of those methods are discussed. 

Classification : Reading and research   Class Number  :  326.803   Credit : 3   Grade : Graduate School

 

Classification : Requisite   Class Number  :  M0000.008700   Credit : 1   Grade : Graduate School

Students are introduced to the faculty and their interests, the field of statistics, computing tips, research ethics and the facilities at the department and the University. Each faculty member gives at least one elementary lecture on some topic of his or her choice. Students are also given information about the e-Learning and Teaching (eTL), the libraries at the University and current bibliographic tools. In addition, students are instructed in the use of the Departmental and University computational facilities and available statistical program packages.

Classification : Requisite   Class Number  :  M0000.008800   Credit : 1   Grade : Graduate School

The statistics department invites experts to Seoul National University to make presentations on the topics of interest in the area of statistics and its applications.

Classification : Requisite   Class Number  :  M1399.000800   Credit : 3   Grade : Graduate School

This seminar course focuses on improving the research, lectures, and related work skills of Ph.D. students in the Department of Statistics. Through lectures, discussions, and seminars from statistics faculty, postdoctoral students, prospective doctoral graduates, and faculty members of the Seoul National University Faculty of Liberal Education, students learn the professional development required for a successful Ph.D. program in statistics. The main contents to be covered during the semester include lecture method seminars to improve lecture skills, statistical research seminars to introduce the latest research topics, and how to write research papers, review reports, research proposals, and resumes.

Classification : Requisite   Class Number  :  M1399.001000   Credit : 3   Grade : Graduate School

This course covers statistical methodologies including data wrangle, data visualization, regression, linear models, generalized linear model, mixed models, and classification, which are essential for master’s level graduate students who are interested in data science. Compared to the traditional courses, this course concentrates on using current systems to do statistics. Less emphasis is made on the underlying theory, which is presented at an intuitive level, and more emphasis is put on using software to implement statistical methods. All statistical analyses of the course are conducted using R and Python.

Classification : Elective   Class Number  :  M1399.001100   Credit : 3   Grade : Graduate School

An Advanced Probabilistic graphical model is a popular framework for encoding joint distributions over complex domains where nodes correspond to variables of interest and the edges of the graph describe conditional dependence information and causal/directional relationships among the variables. Hence, the models are often a foundation of many machine learning approaches for including bioinformatics, social science, image processing, and marketing analysis. This course explains the two basic PGM representations: Bayesian Networks applying a directed graph; and Markov networks applying an undirected graph. The course discusses the theoretical properties of these representations, learning algorithms, as well as their use in practice.

Classification : Elective   Class Number  :  M1399.001200   Credit : 3   Grade : Graduate School

Many questions in empirical scientific research are often causal. This course will introduce statistical theory and methods of causal inference with applications in various scientific fields. Students are expected to understand theoretical frameworks and make valid causal inference. Topics will include potential outcomes; randomized experiments; confounding; observational studies; matching; weighting; propensity score methods; doubly robust; unmeasured confounding bias; sensitivity analysis; instrumental variables; quasi-experimental designs.

Classification : Elective   Class Number  :  M1399.001600   Credit : 3   Grade : Graduate School

This course is a hands-on semester course that enables students to acquire the knowledge, skills, and attitudes necessary for the industry career in the future. Students will experience the application methods and status of data analytics knowledge acquired through hands-on practical training to set up a career path. Practical training is conducted on a full-time basis for more than 4 weeks, and training institutions are limited to institutes that have contracted with Seoul National University. This course can be taken only for graduate students who have completed one semester or more. This course cannot be taken in the semester, when the student is planning to graduate. In addition, other courses cannot be taken when taking this course.

Classification : Elective   Class Number  :  M1399.001700   Credit : 3   Grade : Graduate School

In this course, students undertake a capstone design project to solve real-world problems using data in industry or society based on data analysis practice learned from various statistics courses. Through this data-driven decision making process, students cultivate creativity, professionalism and applied skills to solve practical problems.