ETF3500 - High dimensional data analysis - 2018

6 points, SCA Band 2, 0.125 EFTSL

Undergraduate - Unit

Refer to the specific census and withdrawal dates for the semester(s) in which this unit is offered.

Faculty

Business and Economics

Organisational Unit

Department of Econometrics and Business Statistics

Chief examiner(s)

Dr Anastasios Panagiotelis

Coordinator(s)

Dr Anastasios Panagiotelis

Unit guides

Offered

Caulfield

  • Second semester 2018 (On-campus)

Prerequisites

ETF2100, ETF2121, ETC2410, ETC2420 or equivalent.

Prohibitions

ETC3500

Synopsis

In many fields of business, analysts must deal with data on many variables, for example, surveys with a large number of questions. In such cases, statistical tools known as multivariate methods must be used to analyse the data and drive business decisions.

This unit covers such methods in three sections: Cluster Analysis, Discriminant Analysis and MANOVA can be used to identify, predict and test for differences groups such as between distinct classes of customers or products; Principal Components Analysis, Correspondence Analysis and Multidimensional Scaling are dimension reduction methods that help analysts to visualise complicated datasets; and finally, Factor Analysis and Structural Equation Modelling are used to predict and test theories and explain and predict business outcomes.

Outcomes

The learning goals associated with this unit are to:

  1. demonstrate an understanding of the role that multivariate statistical techniques such as factor analysis, structural equation modelling, categorical data analysis, cluster analysis, multidimensional scaling and correspondence analysis play in uncovering relationships and patterns in survey data
  2. appraise the strengths and limitations of these techniques
  3. apply tools in R to generate solutions for the appropriate statistical techniques
  4. demonstrate skills in using the appropriate statistical techniques from a user and provider perspective
  5. demonstrate skills in communicating the results of the analysis so that decision making can be implemented.

Assessment

Within semester assessment: 50% + Examination: 50%

Workload requirements

Minimum total expected workload to achieve the learning outcomes for this unit is 144 hours per semester typically comprising a mixture of scheduled learning activities and independent study. Independent study may include associated readings, assessment and preparation for scheduled activities. The unit requires on average three/four hours of scheduled activities per week. Scheduled activities may include a combination of teacher directed learning, peer directed learning and online engagement.

See also Unit timetable information