36103 Statistical Thinking for Data Science8cp; block mode with on-campus modules (2 x Saturday workshops, 2 x Monday evenings and 4 x optional clinics) and online work in between; availability: Master of Data Science and Innovation and Master of Business Analytics students
There are course requisites for this subject. See access conditions.
Knowledge of concepts taught in 35513 Statistical Methods is assumed.
This subject helps students to advance their thinking about statistics and how it can be used, or abused, in data science. Starting from the assumed knowledge of basic statistics that students bring into the subject: including concepts like probability, distributions, hypothesis testing, significance, power and confidence; students quickly develop their ability to create modern statistical models in real-world data science contexts. Learning to use the powerful language R, students work their way through the entire data science cycle: from data collection, cleaning and merging datasets, exploratory analysis, modelling and reporting. This process provides rapid exposure to the wide range of modern day packages (for example, the tidyverse) that facilitate rapid statistical analyses for data science questions. Students also learn to make the invisible trends in datasets visible, to make predictions from complex datasets and to reproducibly document their statistical procedures for different audiences. Working with a team of data science professionals from a variety of different backgrounds, students learn how to appropriately communicate their newfound statistical insights and engage a variety of different audiences and stakeholders in order to inform decision making. A selection of advanced topics helps each student to concurrently follow their own personalised learning journey: evaluating and bolstering any gaps in their knowledge and skills, and prepare for future electives and iLab projects.
Autumn session, City campus
Detailed subject description.