This module introduces descriptive and predictive statistics and probability. It covers why statistics is needed and how it is used in data analytics. The students will apply the concepts to several use cases.
The aim of this module is to
-
Introduce the student to the statistical concepts of collecting and analysing data.
-
Give the student skills to create statistical reports through a mixture of theory and practice.
-
Equip the student with the ability to create a theory, as well as build and test hypotheses.
- Fundamentals of using statistics in data analysis
- Populations and samples; Representativeness; Sample Size; Central Limit
- Theorem and its role in data analysis; Sources of bias and dealing with bias
- Different data types; Data distributions and how to recognise and describe them
- Measures of centrality and variance; Role of normality
- Goodness of fit; Fitting a model; Validity and Reliability
- Measuring effect size; Type I and Type II errors; Statistical significance and statistical power
- Conducting Basic Statistical Tests
- Hypothesis Testing; Identifying correct tests to use in a range of analyses
- Understanding and interpreting test statistics
- Tests for correlation, comparing means, analysis of variance; Parametric and non-parametric
- Bayes Theorem
- Discrete and continuous probability: basic concepts and definitions; interpreting probability; different distributions; confidence intervals
- Reporting Analysis
- Structuring a report on data analysis; Descriptive statistics required for different data types
The module explores the theory of descriptive and predictive statistics in lectures, demonstrates practical implementation through R and Python code and teaches the student to implement their own reports in lab sessions.
Teaching of this module will be very practical and oriented around real world data and problems. The goal is for students to understand how statistics is applied to effectively build and test hypotheses around data they are investigating. There will be regular feedback and interaction with students to assure that not only statistical tests can be applied correctly, but understanding of why these tests are necessary and what their interpretations mean is assured as well. The major goals are 1) to prepare them for deeper and more complex data analysis and model development in Machine Learning and 2) demonstrate that statistical analysis can get them very far towards making efficient and reliable predictions with low complexity and high explainability.
| Module Content & Assessment | |
|---|---|
| Assessment Breakdown | % |
| Other Assessment(s) | 100 |