Research Mission
The three major missions of the Statistics and Population Studies Department are:
- To create new scientific knowledge in Statistics, Data Science and Population Studies
- To train graduate students to become active contributors of scientific knowledge
- To impart to undergraduates and postgraduates, the excitement of research through scientific method
As a research university, we believe that active participation in all aspects of the research process provides students with a unique experience: Hands-on learning, deep learning experiences and opportunities that sharpen critical thinking skills.
Our department has interests in many branches of research such as:
- Pure Statistics
- Applied Statistics
- Data Science
- Population Studies
- Computational Finance
~ Ali Ibn Abu Talib ~
"Research is to see what everybody else has seen, and to think what nobody else has thought"
~ Albert Szent-Gyorgyi ~
The Statistics and Population Studies Department revolves around two axes of teaching and research:
- Statistics with specialisation in Data Science
- Population Studies
Indeed, the research interests in the Department are vast, as displayed by the below list of researchers and their research interests.
- Fertility: Infant, child and maternal mortality
- Women empowerment
- Public health: Health policy and reproductive health.
Current Project:
Demographic analysis of inequalities and healthcare services in South Africa- Biostatistics: Data Mining
- Science Education
Current Projects:
- Understanding risk behaviour of entering first-year students.
- The development of graduate attributes in a situated learning environment.
- Entry requirements of university students and their success.
- Describing illness and disability in South Africa.
- Biostatistics: Data Mining
- Science Education
Current Projects:
- Understanding risk behaviour of entering first-year students.
- The development of graduate attributes in a situated learning environment.
- Entry requirements of university students and their success.
- Describing illness and disability in South Africa.
- Mixed modelling with emphasis on Latent Class Models
- Agreement
- Loglinear modeling in contingency tables
- Logistic regression modeling
- Randomised controlled trials
- Design of trials and quasi-experimental studies
- Rasch analysis
- Measurement models
- Multilevel modeling Indirect methods for estimating prevalence
- Survey sampling
- Bootstrap
- Extreme value theory
- Poverty and Inequality.
Current Project:
Statistical analysis of data obtained from complex designs.- Female Genital Mutilation
- Contraception
Current Projects:
Female Genital Mutilation and its demographic characteristics- Two distinct areas of research: First point of interest is reproductive health and contraception questions, and second area of research concentrates on data collection with emphasis on surveys and censuses.
Current Projects:
- Contraception and reproductive health in developing countries
- Census round 2020
- Comparative study of domestic violence in selected countries
- Female Genital Mutilation and its demographic characteristics
- Accessibility of health facilities for young mothers in developing countries
The application of statistical methods to astronomical problems ("astrostatistics"), and the study of variable stars and brown dwarfs.
Decision criteria for choosing between continuous and differential photoelectric photometry of stars, based on time series models.
Fitting time series models to X-ray observations of a black hole low-mass X-ray binary.
Interpretation of spectroscopic measurements of the unusual variable star UNSW V760.
Comparing nonparametric deconvolution of densities by the Lucy (conditional distribution) and characteristic-function methods.
Fitting stochastic differential equations to non-regularly spaced time series.
Extension of a test statistic for changes in a time series, to deal with irregular time spacing, and (ii) complex structure of the series.
Modelling of multivariate dependence using asymmetric copulas.
Parametric deconvolution of densities contaminated by heteroscedastic measurement errors.
Follow-up time series measurements of variable ultracool dwarfs.
Current Projects:
Investigating the efficacy of saddle point methods in the parametric deconvolution of error-contaminated probability density functions.Decision criteria for choosing between continuous and differential photoelectric photometry of stars, based on time series models.
Fitting time series models to X-ray observations of a black hole low-mass X-ray binary.
Interpretation of spectroscopic measurements of the unusual variable star UNSW V760.
Comparing nonparametric deconvolution of densities by the Lucy (conditional distribution) and characteristic-function methods.
Fitting stochastic differential equations to non-regularly spaced time series.
Extension of a test statistic for changes in a time series, to deal with irregular time spacing, and (ii) complex structure of the series.
Modelling of multivariate dependence using asymmetric copulas.
Parametric deconvolution of densities contaminated by heteroscedastic measurement errors.
Follow-up time series measurements of variable ultracool dwarfs.
- Financial Applications in Currency Markets
- Time-related Value Functions a
- Time-varying Regression.
- Various demographic and bio statistical applications have also been studied e.g. spatial analysis, Poisson zero-inflated models and missing value theory for non-ordered categorical data with applications using DHS data.
Current Projects:
- Examining long-run relationships between the BRICS stock market indices to identify opportunities for implementation of statistical arbitrage strategies.
- Using the Markov Chain Monte Carlo method to make inferences on items of data contaminated by missing values.
- Loss given default in the banking environment.
- Influence of associated factors of poverty on educational attainment.
- Multi-year Trend Analysis of Childhood Immunisation Uptake and Coverage.
- Epidemiological analysis of spatially misaligned data.
- Science
- Education
- Biostatistics
- Survival Analysis
- Psychology
- Complex Sampling
- Statistical Modelling
- Prediction Error
- Resampling Methods
- Bootstrap
- Weight Trimming
- Benchmarking
- High Performance Computing
- Data Science
