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:
Statistical objectives
-Prove a result
---eg animals with mutation on gene X have higher levels of Y
---Hypothesis tests, p-values
-Search data for interesting information - ‘Exploratory’
---eg which genes are related to condition Z?
---Generate hypotheses
-Build a predictive model
---Eg predict which individuals are likely to develop a condition based on various factors
---Model to calculate probability of condition
Why Statistical Modelling?
-Simple group comparisons assume groups are independent
-But there may be structure in data : -
---Paired data
---Animals grouped by cages or area
---Repeated experiments
---Data on plates
---Measurements taken over several timepoints
-Question may not be simply ‘Do the groups differ statistically?’
-Other potential objectives :
---Do groups differ statistically after allowing for confounding factors?
---Does the difference vary significantly across areas, experiments, time, etc?
---Can an outcome be predicted from a set of measurements?
General Linear Models (GLMs)
-Simple ANOVA
-ANOVA with more factors (Two-way etc)
-Combination of ANOVA and Regression
-Multiple Regression
-Regression
- Tags
-