SyllabusThe subject will cover the following topics:
- Descriptive statistics: Graphs and statistics as summaries, frequency tables, concept of population and sample, random variation, measurement levels (categorical, ordinal, interval, ratio), measures of location and dispersion, mean and median, standard deviation, range and coefficient of variation.
- Basic probability: Probability estimate from relative frequency and percentiles, conditional probability, probabilities when not all outcomes are equally likely.
- Random variables: Representing the outcome of experiments in terms of random variables, distinguishing between discrete and continuous random variables, calculating probabilities using the Binomial, Poisson Exponential and Normal distributions. Describing and verifying assumptions of Binomial and Poisson distributions.
- Sampling distributions: The application of various sampling strategies and the distribution of sample means and proportions.
- Inference based on a single sample: Using a sample to gain information about a population, distribution of sample means and proportions, confidence intervals, hypothesis tests.Understanding power and what affects it. Introduction to non-parametric tests.
- Inference based on two independent samples: Using samples to gain information about populations, controlled experimentation, distribution of differences in sample means or proportions, confidence intervals, hypothesis tests. Comparison of paired-sample tests with tests for two independent samples.
- Analysis of variance: One-way Analysis of Variance for testing differences in more than two means, graphical displays, checks on assumptions and post hoc tests. Introduction to two-way Analysis of Variance.
- Chi-square tests: Testing whether data fit a specified distribution, checking distributions graphically, and testing for independence of two categorical variables.
- Simple linear regression and correlation: Estimating the intercept and slope of the Least Squares line, testing the slope of the regression line, interpreting coefficient of determination (R squared) and correlation. Introduction to multiple linear regression.
- Experimental design: Difference between experimental and observational data, randomisation, causality, replication, blocking and sample size.
|
|