We are announcing three courses, which are scheduled to take place in October 2004. Summary information is given below.
For more detailed information and registration forms please see http://www.ssc.rdg.ac.uk providing your address and/or fax number.
Using SAS PROC GLM
Dates: 11 - 12 October. Duration: 2 days. Price: £540. (Fee of £1250 for attending this course together with Analysis of Random Effects Models using SAS PROC MIXED (11 - 15 October)).
The GLM procedure in SAS uses the method of least squares to fit general linear models, e.g. simple and multiple linear regression models and general analysis of variance models involving balanced and unbalanced data.
In this course the facilities of SAS PROC GLM are described and assessed. The nature of the four types of sums of squares will be explained, as well as the use of estimating functions. Numerous examples will be used for illustration while the practical sessions, based on SAS output, will help participants understand the ideas involved.
Analysis of Random Effects Models using SAS PROC MIXED
Date: 13 - 15 October. Duration: 3 days. Price: £810. (£1250 for attending Using SAS PROC GLM together with this course (11 -15 October)).
MIXED is a SAS procedure for fitting models with both fixed and random effects. Such models arise when treatments are a random selection from a wider group and when data are collected from a multi-strata structure with different levels of variability.
Practical situations where such experimental data arise are many. For example, in a clinical trial to evaluate the effectiveness of two drugs, the drugs may be administered to patients from a random selection of hospitals with a view to generalising conclusions to patients from all hospitals. In environmental studies, random locations within randomly selected sites may be measured for soil and environmental characteristics. How models for such situations can be fitted in PROC MIXED, and how the output can be interpreted will be covered in lectures and computer practical sessions.
Introduction to Survival Analysis
Dates: 18 - 20 October. Duration: 3 days. Price: £840.
Survival data arise in a literal form from trials concerning life-threatening conditions, but the methodology can also be applied to other waiting times such as the duration of pain relief. This course discusses both the design and analysis of clinical trials in which the response variable is a survival time.
During lectures the statistical package SAS will be used to illustrate the methodologies, and in practical sessions participants will analyse and report on the results of a simulated clinical trial. Considerable emphasis is placed on practical work and the interpretation of models.