CUHK MSC
ASDS
M.Sc. in Advanced Studiesin Statistics and Data Science
OVERVIEW
INTRODUCTION
The program is orgainized by the Department of Statistics. The programme is tailored for students with strong statistics and data science background and interest in further explorations of the subjects. We provide advanced training theoretically and practically and equip the graduates with skills necessary for knowledge discovery and decision making.
FEATURES
Learn advanced theory in
statistics and data science
One-year Full Time or
Two-year Part Time
Equip yourself
lifelong learner
CURRICULUM
MODE OF STUDIES
FULL-TIME MODE
The normative study period of full-time study mode is one year. Students should satisfy the following requirements in order to obtain the degree.
TOTALAt least 24 CREDITS
Note: Students are required to complete three core courses and at least 5 elective courses in order to graduate.
CORE COURSES
STAT 5010
STAT 5010ADVANCED STATISTICAL INFERENCE
This course is concerned with the fundamental theory of statistical inference. Topics include exponential families of distributions, sufficient statistics, convex loss functions, UMVU estimators, estimator performance, information inequality and the principle of equivariance. Bayes estimation, minimax estimation, large-sample comparisons of estimators and asymptotic efficiency will also be covered if time allows.
STAT 5202
STAT 5202STATISTICAL MACHINE LEARNING
This course focuses on statistical theory and methodology in machine learning. It presents theoretical foundation together with intuition and practical aspects to help students develop thorough understanding of modern machine learning methods. The course covers key concepts and statistical theory for machine learning, including supervised learning (regression, classification, etc), unsupervised learning (clustering, dimension reduction, etc), as well as other advanced topics (graphical models, network models, recommender systems, etc).
STAT 5203
STAT 5203APPLIED LINEAR MODELS FOR DATA SCIENCE
This course introduces fundamental elements related to linear statistical models, which covers both theory and applications. Topics include, but not limited to, distribution theory, full rank and non-full-rank linear models, analysis of variance, model adequacy checking; model selection; advanced topics of linear models and applications in modern data science.
ELECTIVE COURSES
STAT 5103
STAT 5103HIGH-DIMENSIONAL DATA ANALYSIS
Beginning with an introduction to the basic knowledge in multivariate and high-dimensional data analysis, including multinormal distribution, descriptive statistics, and graphical displays, this course focuses on the dimensionality reduction methods, which are commonly used in high-dimensional data analysis. Selected topics include principal component analysis, factor analysis and canonical correlation analysis.
STAT 5107
STAT 5107DISCRETE DATA ANALYTICS
This course provides a practically oriented treatment of modern methods for the analysis of categorical data. Topics include analysis of two-way contingency tables, logistic regression, log-linear model, generalized linear model, classification and regression tree method.
STAT 5204
STAT 5204OPTIMIZATION IN DATA SCIENCE
This course offers an introduction to nonlinear optimisation with applications in data science. The theoretical foundation and the fundamental algorithms for unconstrainted and constrained nonlinear optimisation are studied and applied to the supervised learning models.
STAT 5206
STAT 5206DATA MANAGEMENT
This course introduces the basic concepts in Data Management such as normalization. Structured Query Language (SQL) will be used in the course.
STAT 6204
STAT 6204ADVANCED TIME SERIES
This course introduces the advanced methodology in time series analysis and its applications to finance. The use of statistical packages R will be demonstrated.
STAT 6205
STAT 6205STATISTICAL METHODS IN BIOMEDICAL RESEARCH
The course will cover topics that involve the analysis of large-scale data in biomedicine, including genome-wide association analysis, differential expression analysis, clustering analysis, dimension reduction, pathway analysis, and data integration methods.
STAT 6206
STAT 6206BAYESIAN STATISTICS WITH APPLICATIONS IN MACHINE LEARNING
This course covers the fundamentals of a Bayesian approach to machine learning system. The Bayesian approach allows for a unified and consistent treatment for many model-based machine learning techniques. The course covers topics linear Gaussian systems and their useful models and applications, including common regression and classification methods, Gaussian mixture models, hidden Markov models and Kalman filters. Intelligent agents models and active inference will be discussed if time permitted.
STAT 6207
STAT 6207APPLIED DEEP LEARNING
This course provides understanding of practically oriented machine learning and deep learning algorithms. The first part of the course focuses on the theory such as linear classification and deep learning. The second part focuses on applications, such as, recommender systems, generative adversarial networks, and reinforcement learning.
STAT 6211
STAT 6211RESEARCH WORKSHOP
Students are required to conduct a research project on a current topic in Statistics and Data Science. This course will give students insight into how research is carried out in the field of Statistics and Data Science.
OTHER UNIVERSITY REQUIREMENTS
(a) Students much fulfill the Term Assessment Requirement of the Graduate School. For details, please refer to Section 13.0 “Unsatisfactory Performance and Discontinuation of Studies” of the General Regulations Governing Postgraduate Studies which can be accessed from the Graduate School Homepage: http://www.gs.cuhk.edu.hk/.
(b) A student must achieve a cumulative grade point average (GPA) of at least 2.0 in order to graduate.
PART-TIME MODE
The normative study period of part-time study mode is two years. Students should satisfy the following requirements in order to obtain the degree.
TOTAL At least 24 CREDITS
Note: Students are required to complete three core courses and at least 5 elective courses in order to graduate.
CORE COURSES
STAT 5010
STAT 5010ADVANCED STATISTICAL INFERENCE
This course is concerned with the fundamental theory of statistical inference. Topics include exponential families of distributions, sufficient statistics, convex loss functions, UMVU estimators, estimator performance, information inequality and the principle of equivariance. Bayes estimation, minimax estimation, large-sample comparisons of estimators and asymptotic efficiency will also be covered if time allows.
STAT 5202
STAT 5202STATISTICAL MACHINE LEARNING
This course focuses on statistical theory and methodology in machine learning. It presents theoretical foundation together with intuition and practical aspects to help students develop thorough understanding of modern machine learning methods. The course covers key concepts and statistical theory for machine learning, including supervised learning (regression, classification, etc), unsupervised learning (clustering, dimension reduction, etc), as well as other advanced topics (graphical models, network models, recommender systems, etc).
STAT 5203
STAT 5203APPLIED LINEAR MODELS FOR DATA SCIENCE
This course introduces fundamental elements related to linear statistical models, which covers both theory and applications. Topics include, but not limited to, distribution theory, full rank and non-full-rank linear models, analysis of variance, model adequacy checking; model selection; advanced topics of linear models and applications in modern data science.
ELECTIVE COURSES
STAT 5103
STAT 5103HIGH-DIMENSIONAL DATA ANALYSIS
Beginning with an introduction to the basic knowledge in multivariate and high-dimensional data analysis, including multinormal distribution, descriptive statistics, and graphical displays, this course focuses on the dimensionality reduction methods, which are commonly used in high-dimensional data analysis. Selected topics include principal component analysis, factor analysis and canonical correlation analysis.
STAT 5107
STAT 5107DISCRETE DATA ANALYTICS
This course provides a practically oriented treatment of modern methods for the analysis of categorical data. Topics include analysis of two-way contingency tables, logistic regression, log-linear model, generalized linear model, classification and regression tree method.
STAT 5204
STAT 5204OPTIMIZATION IN DATA SCIENCE
This course offers an introduction to nonlinear optimisation with applications in data science. The theoretical foundation and the fundamental algorithms for unconstrainted and constrained nonlinear optimisation are studied and applied to the supervised learning models.
STAT 5206
STAT 5206DATA MANAGEMENT
This course introduces the basic concepts in Data Management such as normalization. Structured Query Language (SQL) will be used in the course.
STAT 6204
STAT 6204ADVANCED TIME SERIES
This course introduces the advanced methodology in time series analysis and its applications to finance. The use of statistical packages R will be demonstrated.
STAT 6205
STAT 6205STATISTICAL METHODS IN BIOMEDICAL RESEARCH
The course will cover topics that involve the analysis of large-scale data in biomedicine, including genome-wide association analysis, differential expression analysis, clustering analysis, dimension reduction, pathway analysis, and data integration methods.
STAT 6206
STAT 6206BAYESIAN STATISTICS WITH APPLICATIONS IN MACHINE LEARNING
This course covers the fundamentals of a Bayesian approach to machine learning system. The Bayesian approach allows for a unified and consistent treatment for many model-based machine learning techniques. The course covers topics linear Gaussian systems and their useful models and applications, including common regression and classification methods, Gaussian mixture models, hidden Markov models and Kalman filters. Intelligent agents models and active inference will be discussed if time permitted.
STAT 6207
STAT 6207APPLIED DEEP LEARNING
This course provides understanding of practically oriented machine learning and deep learning algorithms. The first part of the course focuses on the theory such as linear classification and deep learning. The second part focuses on applications, such as, recommender systems, generative adversarial networks, and reinforcement learning.
STAT 6211
STAT 6211RESEARCH WORKSHOP
Students are required to conduct a research project on a current topic in Statistics and Data Science. This course will give students insight into how research is carried out in the field of Statistics and Data Science.
OTHER UNIVERSITY REQUIREMENTS
(a) Students much fulfill the Term Assessment Requirement of the Graduate School. For details, please refer to Section 13.0 “Unsatisfactory Performance and Discontinuation of Studies” of the General Regulations Governing Postgraduate Studies which can be accessed from the Graduate School Homepage: http://www.gs.cuhk.edu.hk/.
(b) A student must achieve a cumulative grade point average (GPA) of at least 2.0 in order to graduate.
ADMISSION
PROCEDURE
Applicants can submit online applications at Graduate School webpage: https://www.gradsch.cuhk.edu.hk/OnlineApp/login_email.aspx .
Scanned copies of the supporting documents should be uploaded to the Online Application System for Postgraduate Programmes. Official transcripts and hard copies of supporting documents are NOT required at the application stage. For the list of required supporting documents, please refer to Graduate School website https://www.gs.cuhk.edu.hk/admissions/admissions/documents-required .
DEADLINE
The application deadline for 2024/25 admission:
PRIORITY ROUND
31 Dec 2023
FINAL ROUND
28 Feb 2024
Applications will be processed on a rolling basis until all places have been filled. Early applications are strongly encouraged.
TUITION FEE
The tuition fee for 2024/25 admission (subject to University approval):
Part-time Mode: HKD 99,000 per year for two years
Full-time Mode: HKD 198,000 per year
REQUIREMENT
Applicants are required to fulfil the General Admissions Requirements and the English Language Requirements for Admission.
GENERAL ADMISSION REQUIREMENTS
- graduated from a recognized university and obtained a bachelor’s degree in Statistics or related subjects, normally with honours not lower than Second Class or achieving an average grade of not lower than “B”; or
- completed a course of study in a tertiary educational institution and obtained professional or similar qualifications equivalent to an honours degree.
ENGLISH LANGUAGE REQUIREMENTS FOR ADMISSION
To fulfill the University’s minimum English language requirements for admission to postgraduate programmes, applicants should have:
- obtained a degree from a university in Hong Kong1 or taken a degree programme of which the medium of instruction was English; or
- achieved scores in the following English Language tests2 as indicated:
- TOEFL: 550 (Paper-based)/79 (Internet-based);
- IELTS (Academic): 6.5;
- GMAT: Band 21 (Verbal); or
- obtained a pass grade in English in one of the following examinations:
- Hong Kong Advanced Level Examination (AS Level);
- Hong Kong Higher Level Examination;
- CUHK Matriculation Examination;
- General Certificate of Education Examination (GCE) Advanced Level (A-Level)/Advanced Subsidiary Level (AS-Level); or
- achieved Level 4 or above in the English Language subject of the Hong Kong Diploma of Secondary Education (HKDSE) Examination; or
- obtained a recognized professional qualification3, provided that the examination was conducted in English.
Notes:
- This is based on the understanding that English is the medium of instruction of degree programmes offered by universities in Hong Kong. Moreover, graduates from universities in Hong Kong should have fulfilled the English language requirements of the institution concerned when they were admitted to the degree programmes. The CUHK Graduate School may request applicants to provide additional supporting documents to prove their English proficiency.
- TOEFL and IELTS scores are considered valid for two years from the test date. GMAT scores are considered valid for five years from the test date.
- The programmes/Graduate Divisions shall at their discretion to determine if such professional qualification can be accepted.
FAQS
PROSPECTIVE APPLICANTS
This programme is designed for statistics or related subjects graduates. Students with non-quantitative degrees should consider another programme offered by the department.
The candidates are expected to have a solid knowledge of statistics. Knowledges in multivariate calculus, linear algebra and computer programming are also required.
You have fulfilled the English language requirement by obtaining a degree from a university in Hong Kong or a degree in which the medium of instruction was English. A score report of an English Language test is not required. You may, however, choose to provide your score report as supplementary information.
The GRE is not required, but you may choose to submit the report as supplementary information.
Applicants may upload scanned copies of the academic transcripts with an official stamp at the application stage. Official transcripts and hard copies of supporting documents are NOT required at this stage. We will contact applicants for any official documents required.
Yes, electronic transcripts are accepted, and you may arrange for submission by email to (asds@cuhk.edu.hk).
You may amend your information in the application system. However, you will not be able to do so after a certain point in the application stage. If you wish to make amendments, please email us (asds@cuhk.edu.hk) and provide the following information:
- Name
- Programme applied
- Application number
- Identity document number
- Information to be amended
Yes, two confidential recommendation letters are compulsory. However, we do not have specific requirements for the referees. They can be academic or non-academic such as supervisors and colleagues at the workplace.
You should submit the information of your referees including their email addresses in the application system. The system will send an invitation email to your referees at 5:00 am HKT the following day. Your referees may use the link and login information in the emails to complete and submit the recommendation letter. You may check the submission status in the application system, but you will not be able to access the confidential recommendation letter.
Please contact us via email if your referees cannot submit the letters through the application system.
We currently do not have a credit transfer mechanism or course exemption scheme.
Applications are reviewed on a rolling basis. Shortlisted applicants will receive an interview invitation via email or phone.
We do not send individual notifications. Unsuccessful applicants may check their status on the application system in July.
The programme is offered by the Department of Statistics, which is government funded. The programme is not funded by the UGC block grant.
ADMITTED STUDENTS
Please refer to the Finance Office website https://www.fno.cuhk.edu.hk/student/student-fees/others/ for this information.
The programme currently does not offer students loans or financial assistance. Students may explore financial assistance from other units, such as the Extended Non-Means-Tested Loan Scheme (ENLS) administrated by the Student Finance Office.