Introduction to Statistical Modelling
Training session with Dr Helen Brown, Senior Statistician, at The Roslin Institute,  December 2015.
************************************************
These training sessions were given to staff and research students at the Roslin Institute. The material is also used for the Animal Biosciences MSc course taught at the Institute.
************************************************
*Recommended Youtube playback settings for the best viewing experience:  1080p HD
************************************************
Content:
Reasons for using a general linear model GLM
-Can fit more than one effect
---Categorical and continuous effects
-Improve efficiency by :-
---Allowing for data structure, eg cages, plates
---Fitting a covariate, eg pre-intervention measurement
-Control for confounding factors
---Eg Analyse effect of diet on weight after adjustment for age
-Analyse two effects in same analysis
---Eg Diet and breed
-Test for an interaction between effects
---Eg Effect of diet differs between breed?
-Build a predictive model
---Eg predict animal weight from a range of data

GLMs: ‘Dependent’ and ‘Independent’ Variables

GLMs: Key Features
-Suitable when errors (or outcomes) are normally distributed
-Any number of binary, categorical and continuous "xi" variables can be fitted
-Encompasses:
---t-tests
---ANOVA
---Regression
---Analysis of Covariance (ANCOVA)