Background knowledge in statistics equivalent to that provided by LOG708 - Applied Statistics and SPSS

After finishing the course, students should have skills and knowledge in statistical methodology constituting a base from which they can successfully carry out solid empirical work, e.g. in their master thesis or later in their academic or other professional careers. Specifically, the students should be able to

• confidently perform estimation and testing of hypotheses about main population parameters such as means, proportions, variances
• specify and estimate linear regression models, using appropriate theory and sample data
• identify and handle nonlinear effects in regression models using transformations and dummy variables
• identify and handle heteroscedasticity, multicollinearity and autocorrelation in regression data
• work with goodness-of-fit tests, analysis of contingency tables and basic nonparametric methods
• work with basic time series models and do forecasting with moving averages, exponential smoothing
• interpret the result of statistical analyses and explain the results in nontechnical language

The content can be somewhat variable. The core topics are

• Multiple Regression Analysis - basic theory and practical aspects
• Topics in regression - categorical data, nonlinear models, deviations from standard assumptions: specification bias, heteroscedasticity, multicollinearity, autocorrelation
• Nonparametric methods
• goodness-of-fit tests
• Time Series Analysis - decomposition, moving averages, exponential smoothing, autoregressive models

Additional topics may include

• Discrete response regression models
• Factor analysis
• Validity

Newbold, Carlson and Thorne, Statistics for Business and Economics, 7. edition, Pearson, Chapter 11 - 14, 16