Workshop 1: Spatio-Temporal Statistics with R
Presenters:
Distinguished Professor Christopher K. Wikle, University of Missouri
Dr Petra Kuhnert, CSIRO Canberra
Date: Saturday 25th November & Sunday 26th November
Workshop Outline:
The workshop gives a contemporary presentation of spatio-temporal processes and data analysis, bridging classic ideas with modern hierarchical statistical modeling concepts. From understanding environmental and ecological processes to developing new technologies for mapping public-health data and the spread of invasive-species, there is a high demand for statistical analyses of data that take spatial, temporal, and spatio-temporal information into account. This course presents a systematic approach to key quantitative techniques for the statistical analysis of such data that features visualization and exploration, covariance-based and random effects models, and dynamic spatio-temporal models. Many examples will be included, along with some basic applications from various R packages.
Recommendations and Prerequisites:
The workshop assumes that participants have an understanding of regression from the matrix algebra perspective, mixed models, and the notion of likelihoods.
Workshop 2: Methodological aspects of the use of geospatial technology for agriculture and agri-environmental statistics
Presenter: Professor Elisabetta Carfagna, University of Bologna
Date: Sunday 26th November
Workshop Outline:
This workshop addresses the main methodological aspects behind the use of geospatial technology for producing reliable and timely agriculture and agri-environmental statistics. Different kinds of geospatial technology are taken into consideration and advantages, constraints and requirements are highlighted. Various statistical aspects are taken into consideration, such as main types of agricultural and agro-environmental probability sample surveys based on area sampling frames and corresponding estimation methods, the use of remote sensing data at the design level (area frame construction and stratification) as well as at the estimator level. Finally, the impact on estimates of spatial resolution, change of support and transformations of spatial data is addressed.
An outline of the workshop is as follows:
- Agricultural and agri-environmental probability sample surveys for exploiting geospatial information
- List and area sampling frames
- Area frame construction and stratification
- Optimisation of area sample designs
- Sample allocation
- Multiple frame sample surveys
- Master sampling frames
- Estimation methods using geospatial information
- Associating segments with reporting units
- Sampling circles and farms by points and related estimators
- Multiple frame estimators
- Improving the precision of estimates with geospatial auxiliary variables
- Improving the precision of estimates with geospatial auxiliary variables – regression and calibrating estimators
- Small area estimators
- Impact on estimates of spatial resolution, change of support and transformations of spatial data
Recommendations and Prerequisites:
The workshop assumes that participants have a basic knowledge of survey methodology.
Workshop 3: Exploring data and models visually
Presenter: Professor Di Cook, Monash University
Date: Sunday 26th November
Workshop Outline:
- Motivation, why and how to organise data and getting started with R:
- Participants get started with R, learn how to organise a work project and use knitr to incorporate code into documentation to produce pdf, html or Word documents.
- Concepts of tidy data and learning to rearrange data will be covered.
- Making basic plots, grammar of graphics, good practices:
- Mapping data to graphical elements in plots using ggplot2.
- Generating simple plots, scatterplots, bar charts, time series, profiles, boxplots.
- Using cognitive principles to improve plots.
- Advanced graphics, layering, maps, interactivity:
- Layering different data sets, drawing maps, exploring model fits, multivariate plots.
- Simple interactive graphics.
Recommendations and Prerequisites:
Participants should have a beginner’s knowledge of R. Two optional readings before the workshop are:
- Cookbook for R, by Winston Chang
- R for Data Science, by Garret Grolemund & Hadley Wickham