ABSTRACT
We open with the distinction between correlation and regression before explaining the calculation and interpretation of the correlation coefficient, r, and the coefficient of determination, R-squared. We introduce simple linear regression, distinguishing it from non-linear and multiple regression. We clarify the components of regression models, the intercept and the slope, before demonstrating how to find the line of best fit using least squares regression, plotting it and using it to predict expected values of the dependent variable, Y, for specific values of the independent variable, X. We proceed to cover index numbers in the context of measuring changes in price levels, and how they are compiled, from the simple aggregate price index to Laspeyres and Paasche price indexes. We show how measures of price movements, such as the Retail Price Index in the UK, are used to ‘deflate’ or adjust money amounts for the effect of inflation across time. We introduce time series analysis, defining the components of time series in the classical decomposition technique; trend, recurrent (weekly, monthly, seasonal and cyclical) and error. We demonstrate the use of moving averages and, where appropriate, centring, in breaking down a time series into its components before showing how to use the model to construct forecasts of future values.
