Module Descriptors

Business

MA218: Advanced Statistical Methods for Business (5 ECTS)

This module demontsrates further applications in inferential statistics with applications in Economics, Business, Marketing and Finance.

Students are required to have completed an introductory course in Descriptive Statistics and Probability equivalent to MA109 Statistics for Business and an introductory course in Inferential Statistics, equivalent to MA217 Statistical Methods for Business.

Lectures will be used to present the ideas of statistical theory and practice, these will include demonstrations of real-life statistical analyses based on business oriented problems. Students will be required to apply the methodology to example datasets using suitable statistical software, SPSS. The tutorials and practical sessions will give support for this aspect of the course.

Taught in Semester(s) II. Examined in Semester(s) II.

Workload: 34 hours (24 Lecture hours, 10 Lab hours).

Module Learning Outcomes. On successful completion of this module the learner should be able to:

(Not all topics described below may be covered in any one academic year, topics covered can depend on students' interests)

  1. describe the conditions when a parametric test or a non-parametric test alternative is more suitable in certain problems and complete non-parametric tesing procedures: the Sign Test, the Wilcoxon Rank Sum Test, the Wilcoxon Signed Rank Test.
  2. complete advanced analysis of simple linear regression models, including analysis if variance (ANOVA) in the response variable, calculation of R-squared and completion of the ANOVA table, further exploration into inference for regression model parameters, including individual t-tests for model parameters and the F-test through calculation of ANOVA table, check assumptions, diagnostic and model checking, apply transformations, using the model for prediction, prediction intervals and confidence intervals.
  3. carry out analysis for problems requiring a multiple regression model, i.e. many candidate predictors may be used to exlain the variability in the response variable. This includes, interpretation of regression coefficients, constructing confidence intervals and carrying out hypothesis tests for regression parameters through individual t-tests and the ANOVA F-test, using the model for prediction, prediction intervals and confidence intervals, model -buidling,variable selection procedures, adjusted R-squared.
  4. extend knowledge of multiple regression to interpretation of model coefficients for models with qualitative/categorical predictors and analysis of covariance, ANCOVA.
  5. complete basic time series analysis, including descriptive analysis via calculation of various index numbers, recognise time series components, apply smoothing models such as Moving-average models, Exponential Smoothing, and Holt-Winters Smoothing, apply seasonal regression models using additive models with indicator variables, describe autocorrelation and carry out the Durbin-Watson test, use models to forecasting trends and seasonality, measure Forecasting accuracy using MAD, MAPE, and RMSE.
  6. apply methods in Quality Control including compliation and interpretation of statistical control charts for monitoring the mean of a process, monitoring the variation of a process, and for monitoring the proportion of defectives generated by a process, diagnosing the cause of variation and capability analysis.
  7. to communicate results of analyses in clear, structured reports.

Indicative Content

Module Resources

Back