This session explores some of the most widely known modelling techniques. We will discuss when regression methods are appropriate and how to understand and select the variables that go into your model. The difference between data types will be discussed and how this influences the model will be described. While linear and binomial logistic regression models are the most commonly used, other models will be discussed.
Independent Variables, Dependent Variables, Covariates, Confounders, Correlation, scatterplots, line of best fit (least squares), simple linear regression, multiple linear regression, multicollinearity, autocorrelation, heteroskedasticity data transformation for non-linear relationships, interactions between variables, forward selection, backward elimination and stepwise approaches, logistic regression, exp(B) and odds ratios, ordinal and multinomial logistic regression, polynomial regression