Introduction to Statistical Modelling and Flexible Models for Correlated Count or Multi-Category data Workshops

12 – 15 January 2016 and 18-21 January 2016

The Centre for Research into Ecological and Environmental Modelling (CREEM) will be running two Statistical Modelling workshops to introduce both basic and more advanced modelling methods. The first workshop is a four day introductory course and the second a four day workshop on flexible models for correlated count or multi-category data.

The courses will take place in our purpose-built facilities at the University of St Andrews, Scotland.  Both courses will be heavily data (and practical) based and the methods will be illustrated using large scale offshore wind farm data. The R software package will be used for the workshop-based practical’s. Computer sessions take place in our computer classroom (attached to the seminar room) and participants can use our computers or bring their own laptops. If you are not familiar with R, we suggest this online tutorial before attending (R tutorial).

The content of the courses are below:

1) Introduction to Statistical Modelling (4 days)

This workshop will take place 12th – 15th January 2016.  This 4 day course is taught using methods that can be widely applied but is illustrated using an Enivironmental Impact Assessment case study (using a large scale offshore windfarm). This course starts at a very basic level and covers the following:

  • Exploratory data analysis: graphical illustrations (e.g. histograms, boxplots etc)
  • Summary statistics, the basics of parameter estimation and inference (e.g. means, medians, variances)
  • Confidence Intervals, Hypothesis Testing (including t-tests and ANOVAs)
  • Models for continuous response data:
    • Multiple linear models: including model specification, fitting, selection, assessment and interpretation (with a particular focus on interpretation)
  • Models for count data and models for binary responses (e.g. presence/absence data):
    • Generalized Linear Models (GLMs): including model specification, selection, fitting, diagnostics and interpretation. Accounting for overdispersion will also be covered.
  • Models for correlated count data and correlated binary data: Generalized Estimating Equations (GEEs).

2) Flexible Models for Correlated Count or Multi-Category data (4 days)

This workshop will take place 18th – 21st January 2016 and covers one dimensional and two dimensional smoothing using more traditional and recently developed methods. The correlated aspect of the data is addressed using Generalised Estimating Equations (GEEs). These will be covered in more detail than in the introductory workshop.  The initial example will be using count data (Poisson) however towards the end of the workshop we will also cover models for categorical data (e.g. Binomial, Binary and Multinomial response types).

———–

To help you decide what workshops to attend, we have prepared a small quiz for you to assess your knowledge of the material covered in the introductory workshop.  The link for the quiz is here (login: STANDLSH) .

Schedule
The workshops will begin at 09:00 and end at 17:00 each day, with registration at CREEM from 08:45 on the first day.

Registration Fees
The number of participants on each workshop will be restricted to 35. Places will be filled on a first-come, first served basis on the receipt of a completed registration form and payment in full.

Workshop 1 – Introduction to Statistical Modelling (4 days)
Fee – 600.00 GBP, for a regular participant (500.00 GBP, for a registered student)

After 14th December the registration fee will increase to:

Fee – 650.00 GBP, for a regular participant (550.00 GBP, for a registered student)

Workshop 2 – Flexible Models for Correlated Count or Multi-Category data (4 days)
Fee – 600.00 GBP, for a regular participant (500.00 GBP, for a registered student)

After 14th December the registration fee will increase to:

Fee – 650.00 GBP, for a regular participant (550.00 GBP, for a registered student)

Workshops 1 & 2 – attending both workshops
Fee – 1100.00 GBP, for a regular participant (1,000.00 GBP, for a registered student)

After 14th December the registration fee will increase to:

Fee – 1,200.00 GBP, for a regular participant (1,100.00 GBP, for a registered student)

Please complete the registration form below and submit:

Introduction to Statistical Modelling (Workshop 1) and Flexible Models for Correlated Count or Multi-Category Data (Workshop 2) Registration

  • Please complete your details below. Fields marked * are required
 

Verification

Fees cover tuition, all course materials, tea/coffee breaks and lunches.

Payment
Payment is made via the University Online Payment Service, please follow this link to the payment site.

These workshops subsidise our charitable teaching of statistical modelling in Africa. In previous years, workshops have been held in Kenya, Mozambique, South Africa (University of Witwatersrand, University of Limpopo, University of Cape Town, Kruger National Park) and Ghana.

Location and Travel
Information is available here.

Accommodation
Accommodation information is available here.

Enquiries
For information please contact Rhona Rodger at:
Centre for Research into Ecological and Environmental Modelling
University of St Andrews
The Observatory, Buchanan Gardens
St. Andrews
Scotland KY16 9LZ
Tel:+44 1334 461842
Fax: +44 1334 461800
Email: rmr5@st-andrews.ac.uk