Correlation coefficient, linear regression, least squares estimation of parameters, unbiasedness, minimum variance properties of estimators, Gauss-Markov theorem, tests of hypotheses concerning parameters of regression model, prediction, analysis of variance, coefficient of determination. Multiple linear regression, polynomial regression, and time series models. Components of time series data. Fitting trend line. Least squares estimators of regression coefficients using matrix notation. Inferences concerning regression coefficients, choice of a fitted model through hypothesis testing. Computer implementations using appropriate statistical software and calculators.
Contact hours:
Four lectures/tutorials and one computer lab per week.Assessment:
Two tests (15%), approximately four assignments (15%), project (10%), final exam (60%).Prerequisite: 1.30812 and 1.20804
Text:
Draper, N. R. & Smith, H., 1981, Applied Regression Analysis, John Wiley, New York
Neter, J., Wasserman, W. & Kutner, M.H., 1983, Applied Linear Regression Models, Richard D. Irwin, Illinois