Geographically Weighted Regression: Two-day training workshop
Registration for this course is now closed.
presented by Stewart Fotheringham,
Science Foundation Ireland Research Professor and Director of National Centre for Geocomputation at NUI Maynooth
and Dr Urška Demšar, National Centre for Geocomputation, NUI Maynooth
Venue: University of Dundee
Course Content
This two-day course introduces the technique of Geographically Weighted Regression (GWR) for modelling spatially-varying relationships in social science data. The course consists of lectures, examples and practical sessions using GWR and ArcGIS.
Issues to be covered will include the following:
- review of traditional regression
- differences between GWR and conventional linear regression models
- elements of a GWR model
- requirements for running a GWR model
- interpreting GWR model output
- extensions to the basic model
Examples will include sample data sets as well as analyses from recent papers from research areas in the social sciences. As the course proceeds, the methods introduced will use more complex material, but presentations will focus on use and interpretation rather than mathematical detail.
Who should attend?
The course is intended for those using or considering using regression modelling or relative multivariate methods within social research contexts.
Why attend?
GWR is essential to all those involved in investigating relationships within spatial data. It is applicable to any form of analysis in which a dependent variable is specified and its distribution is examined with respect to a series of independent covariates. GWR adds powerfully to our ability to understand potential spatial variations in processes and allows us to investigate a "Geography of Relationships".
Course Objectives
- to introduce participants to appropriate use of GWR.
- to understand the structure of a GWR model and how this differs from a conventional regression model
- to understand the choice and how to select between different spatial weighting methods
- to understand how to use GWR software
- to give experience interpreting GWR model output.
Learning Outcomes
By attending this course you will:
- gain an understanding of the principles of the GWR approach, underlying assumptions and limitations, necessary pre-processing/data structuring, model options
- be able to understand model results
- be able to choose between different weights and options
- understand how GWR can be useful in your research
Prerequisites
The course will be pitched at an intermediate level. Participants must be competent using linear regression in social research contexts. All participants must have sufficient elementary algebra to understand weights matrices. It would be desirable to have had some exposure to GIS and spatial data analysis.
Application Process
There are 15 places available on this course. Places will be allocated following a process of application and priority will be given to students and academic staff in Scotland who meet the selection criteria and for which the course is free to attend.
Non-academics and academics from outside Scotland may apply but there will be a charge to attend as follows:
- Academic students and staff in Scotland - FREE
- Students and Academic staff outside Scotland - £100 (£50 per day)
- Public and Voluntary Sector - £150 (£75 per day)
- Private Sector - £200 (£100 per day)
If you are interested in applying for this course, please complete and submit an application form at the following link:
Application form
Deadline for completed applications is 5pm on Friday 30th September 2011.
Successful applicants will be notified by 5pm on Friday 21st October 2011.
PhD Expenses
PhD students studying in Scotland who live outside Dundee are eligible to claim for travel to and from the event and two nights evening meal and accommodation (7-9 December) where required.
AQMeN will only reimburse travel to/from an address in Scotland as provided on the application form. Travel costs must be kept to a minimum.
AQMeN will reimburse a maximum of £50 per night towards accommodation and up to £15 (excluding alcohol) for evening meals on the 7th and 8th December only.
In order to be reimbursed for travel, meals and/or accommodation, original travel tickets and receipts clearing showing the date and journey taken must be provided. For accommodation and meals a fully itemised receipt MUST be provided. *Credit card receipts and email booking confirmations will not be accepted by our auditors and failure to provide the appropriate documentation will result in us being unable to reimburse you.
Non-attendance
All AQMeN courses are offered free of charge to PhD students and academics in Scotland; however, the courses are expensive to run, both in terms of time and money, and non-attendance on courses impacts on our ability to complete the programme of training and to adequately evaluate the value of these events. This is detrimental to the future sustainability of AQMeN as the funding bodies expect full evaluation.
By applying for this course you are making an undertaking to attend the full two days of the course. Failure to do this may result in you being charged a fee to cover the cost of the course and any accommodation that has been booked. If after a place is allocated and confirmed you are no longer able to attend, you must notify AQMeN as soon as possible to allow the place to be offered to someone on the waiting list. Failure to provide advance notification, resulting in the place not being available to another may result in you being charged a fee to cover the cost of the course.
