Teaching Responsibility
LJMU Schools involved in Delivery:
Computer Science and Mathematics
Learning Methods
Lecture
Practical
Module Offerings
7502BDSA-SEP-PAR
Aims
The course covers the concept, the underlying assumptions, and the applications of generalized linear model (GLM) and growth curves. It considers detecting and handling influential observations. The course provides students with a comprehensive knowledge about matrix representation of a model, model fit, model validation, interpretation and prediction for future observations. The course covers the experimental design and analysis, the analysis of variance (ANOVA), logistic regression, and the analysis of multidimensional contingency tables. The students will have hands-on practice using R/SPSS (or alternative recent software) as analytic tools to analyse real-world data by implementing the course topics.
Learning Outcomes
1.
Model relationships between variables using regression analysis, classification and non-parametric methods.
2.
Implement different resampling methods for model assessment, model selection, and uncertainty measurement.
3.
Control prediction variance using subset selection, shrinkage, and dimension reduction methods.
4.
Implement tree-based methods for decision-making problems.
5.
Analyse real data using advanced statistical programming software.
Module Content
Outline Syllabus:Statistical Learning
Linear Regression
Classification.
Resampling Methods
Linear Model Selection and Regularization
Moving Beyond Linearity
Tree-Based Methods
Additional Information:The module contributes to the master’s aim to equip the student with the required abilities and skills to perform data science on real-world applications.