Skip to main content

Statistical Inference

Overview

  • Credit value: 15 credits at Level 5
  • Convenor: Dr Richard Pymar
  • Assessment: coursework (20%), statistical reports (20%) and a written examination (60%)

Module description

In this module we introduce you to the foundations of statistical inference including hypothesis testing, constructing confidence intervals and non-parametric methods. For students planning to specialise in the statistical aspects of data science, the module provides the essential foundations for further study in this area. Numerous examples will be provided, many of which will explored with the aid of an appropriate statistical programming language and package.

Indicative syllabus

  • Introduction to a statistical computer package (R or some other)
  • Discrete random variables and probability distributions
  • Continuous random variables and probability distributions, including the normal distribution
  • Populations, samples and sampling distributions, including the student’s t, and chi- square, distributions
  • Estimation, confidence intervals and hypothesis testing with special reference to samples from normal distributions and estimates of proportions
  • Chi-square tests of goodness of fit
  • Hypothesis tests in R

Learning objectives

By the end of this module, you will be able to:

  • specify the characteristics, and various examples, of discrete and continuous probability distributions
  • carry out calculations of probabilities for normally distributed random variables (i) by the use of statistical tables or (ii) by the use of a statistical package
  • find estimates and construct confidence intervals for unknown parameter values (i) by the use of statistical tables or (ii) by the use of a statistical package, especially for data from normal distributions and for proportions
  • formulate statistical hypotheses, compute appropriate test statistics, evaluate their significance, (i) by the use of statistical tables or (ii) by the use of a statistical package, and draw conclusions, especially for data from normal distributions and for proportions
  • report clearly and simply the results of statistical analyses in a way that may be understood by non-specialists
  • edit the output from a statistical computer package and incorporate extracts into a word-processed report.