For information on winter intensives that are being delivered partially or fully on campus, please refer to the COVID-19 page.
|Fees||Look up fees|
The aim of this subject is to help students develop capability to effectively summarise environmental variables met in the course of research and design, to select appropriate statistical models describing the data structure, and to conduct statistical inference on underlying processes. Students will apply a variety of models from a conventional or Bayesian approach to solve the problems at hand and derive deterministic or stochastic inferences from them.
The subject is composed of four wide-ranging topics from exploratory data analysis to spatial modelling. At the beginning of each topic, students are provided with a set of data from environmental research, and a number of analysis tools are conveyed in the lectures. The mathematical aspects of the subject build on concepts developed in fundamental engineering mathematics and statistics courses from undergraduate courses. It supports student learning in the capstone design and research projects where data analysis skills are assumed.
The subject provides a fundamental skill for a career in environmental engineering where the ability to analyse and communicate the meaning of time series and spatial data sets are expected.
Specific topics include:
1. Exploratory Data Analysis
- Summary statistics and probability models
- Analysis of variability and hypothesis test
- Linear regression and verification/validation.
2. Time Series Analysis
- Introduction to multivariate analysis
- Principle component analysis
- Stochastic forecast and verification.
3. Methods for Multivariate Data
- Multivariate linear regression
- Principle component analysis.
4. Analysis of Spatial Data
- Simple spatial interpolations
- Analysis of spatial variability
- Spatial models and Kriging.
Intended learning outcomes
INTENDED LEARNING OUTCOMES (ILO)
On completion of this subject the student is expected to:
- Effectively summarise their analysis and design outputs
- Use stochastic approach to make statistical inference about random environmental variables
- Define and evaluate objective functions for their design target
- Quantitatively test their hypothesis
- Select the most appropriate statistical model describing the data at hand
- Generate both deterministic and stochastic realisations of environmental variables.
- Ability to apply knowledge of science and engineering fundamentals
- Ability to undertake problem identification, formulation, and solution
- Proficiency in engineering design
- Ability to conduct an engineering project.
Last updated: 29 April 2020