This document is for Coventry University students for their own use in completing theirassessed work for this module and should not be passed to third parties or posted on anywebsite. Any infringements of this rule should be reported to[email protected]Faculty of Engineering, Environment and Computing7089CEM: Introduction to Statistical Methods forData ScienceAssignment Brief Module TitleIntroduction to Statistical Methods forData ScienceIndividualCohort (Sept/Jan):Sept and Jan startModule Code7089CEMCoursework TitleModelling EEG signals using polynomial regressionHand out date:17/05/2021LecturerDr Fei HeDue date and time:13/08/2021, 18:00Estimated Time (hrs): 4 weeksWord Limit*: 3000 – 4000Coursework type:Individual assignment% of Module Mark: 100%• Submission arrangement: online via Aula, and Turnitin.• File types and method of recording: Report (Word), Programme code (R, or Python/Matlab script)• Mark and Feedback date: 2 weeks after submission• Mark and Feedback method (e.g. in lecture, written via Gradebook): provided in Turnitin Module Learning Outcomes Assessed:• Demonstrate knowledge of underlying concepts in probability and statistics used in DataScience.• Select and apply appropriate statistical methods or techniques to solve problems or analysedata sets.• Use modern software to solve real world problems and analyse large data sets.• Interpret the results of their analyses and communicate those results accurately.Task and Mark distribution:Coursework Description:The aim of this assignment is to select the best regression model (from a candidate set of nonlinearregression models) that can well describe the relationship between several ‘simulated’ brainelectroencephalogram (EEG) signals. EEG is a widely used non-invasive method to record electricalactivity of brain. Compared with other neuroimage data, such as functional magnetic resonance imaging(fMRI), EEG technique is much cheaper and has excellent temporal resolution. As a result, EEG-basedanalysis and modelling approaches have been extensively applied to characterise various neurologicaldisorders (e.g. Parkinson’s disease, epilepsy, tremor) and the development of brain-computer interface.To achieve those goals, it is very important to understand the connectivity between different brain areas,which can be obtained through the modelling and analysis of different EEG channels.Data: This document is for Coventry University students for their own use in completing theirassessed work for this module and should not be passed to third parties or posted on anywebsite. Any infringements of this rule should be reported to[email protected] The ‘simulated’ 5 EEG time-series data are provided in the two separate excel files. The first X.csv filecontains 4 input EEG signals
