This course aims to develop the computational thinking skills required for statistics majors. i.e., the ability to solve problems expressed in logic and to write computer programs logically. In particular, we will study the basic concepts of programming and programming languages, such as data structures, abstraction, hierarchy, modularity, iteration, recursion, procedural thinking, value-oriented thinking, reuse, computational complexity, and data types. A specific programming language (e.g., R) may be employed, but only as a tool to materialize concepts; the material covered is language-neutral. Additional attention is paid to functional programming and object-oriented programming paradigms. Fundamentals of data wrangling, manipulation, and exploration for data analysis are also covered, as well as how to visualize, present, and communicate trends in various data types.

Probabilistic methods are being used to understand almost all phenomena in modern society, not only in the natural sciences. Probability theory is an important branch of modern mathematics and has a wide range of applications in computer science such as artificial intelligence and computer communication. In this course, students will first understand the basic concepts of probability, and then study probabilistic thinking and approaches used in natural, engineering, and social sciences, and introduce the mathematical techniques required for them. It is also a good preparation for Mathematical Statistics 1.

This course considers basic sample design and estimation theory to cover diverse and practical designs. The main contents include simple random sampling, stratified sampling, cluster sampling, stratified multistage sampling, ratio and regression estimation, and non-sample error. Each method covers parameter estimation, sample size estimation, optimal sample allocation, and relative efficiency. Sample design and survey practice are conducted in the field as case studies.

This course introduces the definition of probability and the concepts of distribution functions and probability density functions, which are fundamental to statistics. You will be introduced to various types of distribution functions and their properties. Students will also learn about the concept of a sampling distribution, the distribution of statistic, various properties of sampling distributions and approximations to sampling distributions.

This course covers the basic theory of statistical inference, namely, estimation and testing. Estimation methods include method of moment, maximum likelihood estimation, Bayesian estimation, and minimum variance unbiased estimation, while testing methods include the likelihood ratio test, Rao test, Wald test, Bayesian test, and uniformly most powerful test. This course also covers the theoretical background of statistical inference which includes sufficient statistics, the Rao-Blackwell theorem, the Cramer-Rao inequality, derivation of asymptotic properties of estimators, and asymptotic approximation of tests.

This course basically deals with fitting linear regression models to datasets. It also covers statistical inference on model parameters. The topics include simple linear regression, multiple linear regression, checking model adequacy, weighted least squares, transformation, regression diagnostics, detection of leverage and influential observations, regression techniques for categorical variables, multicollinearity, ridge regression, variable selection, nonlinear regression, generalized linear models, artificial neural networks.

This course focuses on understanding and analysis of categorical data. Categorical data is widely collected in routine surveys and experiments in various fields such as social sciences and public health. Students will learn a variety of analysis methods, including the analysis of contingency tables, generalized linear models, logistic regression models, and log-linear models. Through this course, they will develop the ability to understand the characteristics of data, choose and apply appropriate analysis methods. Moreover, the course covers the application and implementation of these methods on real data through statistical software.

This course introduces several effective methods to collect and analyze data to improve processes, reduce costs, and understand a complex system for both scientific research and industrial applications. The course covers complete randomized design, Latin square design, factorial design, block design, fractional factorial design, and response surface analysis. Students will also learn how to analyze data collected according to each design method. A prerequisite for this course is ‘Regression Analysis and Lab.’.

This course provides the theory and applications of multivariate statistical analysis, aiming to equip students with the ability to effectively analyze and interpret multivariate data. The curriculum covers key topics such as the multivariate normal distribution and related distribution theories, as well as multivariate hypothesis testing for group comparisons and statistical significance analysis. Additionally, students will learn various data analysis techniques, including Principal Component Analysis (PCA), and explore latent variable structures through Factor Analysis. The course also introduces classification methods for pattern recognition and predictive modeling. Through theoretical learning and practical applications, students will develop proficiency in applying multivariate analysis techniques to real-world data.

Bayesian inference is a statistical inference method that represents information with uncertainty as a probability, updates the information with Bayes theorem and make inference using the posterior distribution. This course covers the historical background of Bayesian inference, the basic elements of Bayesian inference, and the Bayesian computational methods needed for practical Bayesian inference, including Markov chain Monte Carlo and variational methods. This course also deals with how to apply Bayesian inference to basic statistical models such as two-sample tests, linear regression, and logistic regression.

This course delves into the concept of data literacy and the various methodologies employed in data mining. It is structured into three parts. The first part focuses on understanding data literacy, highlighting its significance in today’s world and covering essential aspects of data collection, processing, and interpretation. The second part introduces a range of data mining methodologies, emphasizing the comprehension of data-driven relationships from correlation to causation. It also covers the construction of predictive models, from linear regression to neural networks, techniques for model interpretation, and exploratory data analysis methods such as clustering and association analysis. The final part involves practical application, where students analyze real-world examples and present their findings through team projects. The course comprises two hours of theoretical instruction and two hours of practical sessions each week. A prerequisite for this course is ‘Regression Analysis and Lab.’.

This course covers time series analysis methods and the use of statistical packages. It includes topics such as the properties of ARIMA models, seasonal ARIMA models, GARCH models, integer-valued time series models, and inferences for these models, including change point analysis.

In this course, students learn statistical methods to assess abnormality from various forms of data observed over time. The course covers univariate/multivariate Shewhart control charts, CUSUM control charts, EWMA control charts, and non-parametric control charts for both numerical or quantitative data. Some courses that may be helpful include <Multivariate Data Analysis and Lab.>, <Mathematical Statistics 1>, <Mathematical Statistics 2>, <Nonparametric Statistics and Lab.>, and <Time Series Analysis and Lab.>

This course covers both discrete and continuous Markov chains, including topics like recurrence, ergodicity, reversibility, absorptions, and their applications, such as MCMC. It also delves into renewal processes, encompassing Poisson processes, various renewal theorems, and their applications.

The course provides students with an understanding and application of various data analysis methodologies for solving real-world problems. It covers key concepts including problem formulation, data collection and preprocessing, exploratory data analysis, visualization, statistical inference, prediction, and decision-making. In addtion to structured data, the course also covers methods of analyzing unstructured data, such as text data, geospatial data, and network data, along with appropriate visualization and analytic techniques. Students will develop the ability to understand various data structures, select appropriate analysis methods, and apply their knowledge to real-world problems through hands-on practice. This course aims to effectively enhance students’ problem solving skills and practical analytical competencies applicable in professional settings.

This course aims to teach you about various advanced statistical methods used to analyze data. We will extend the theory and application of linear models introduced in regression analysis to more general cases and discuss the intuitive explanations and limitations of each method. You will also learn how to implement these methods through programming to apply them to real-world problems. After completing this course, students should be able to (1) select appropriate statistical analysis methods when faced with new data analysis problems, (2) implement these methods using statistical software or through programming, and (3) explain the results of their analysis to non-statisticians.

Statistics provides the theoretical foundation for rational decision-making based on data. This course covers the concepts of probability and random variables, the basis of statistical theory, as well as statistical estimation and testing. Through this course, students will be able to understand the process of data generation as a probability model and will learn the theoretical foundation for statistical inference methods for data analysis.

※ Students who are minor in statistics can take the course. Students majoring in statistics and students who are double major in statistics are not allowed to take the course.