Experts directory

Prof E Ranganai

College of Science, Engineering and Technology
School of Science
Department: Statistics
Professor
Tel:
011 670 9257
E-mail:
rangae@unisa.ac.za

Expertise

  • Applied Statistics
  • Distribution theory
  • Energy (including renewable) forecasting
  • Quantile Regression
  • Regression Diagnostics
  • Time Series Analysis

Qualifications

  • PhD (Statistics), MSc (Statistics)
  • BSc Honours (Statistics)
  • BSc General (Mathematics & Statistics)

NRF Rating

C2

Currently teaching

  • STA3703: Distribution Theory III
  • STA4804: Regression
  • STA4807: Time Series

Fields of academic interests

  • Variable Selection and Regularization in Quantile Regression as well as other methodological developments in Quantile Regression.
  • Quantile Regression Diagnostics and Regression Diagnostics in General.
  • Time series: Probabilistic load and renewable energy (solar and wind) forecasting including the optimization of grid integration of renewable energies and value at risk (VAR) as well as methodological developments in Time Series.
  • Analysis of Count Time Series Data.
  • Digital Financial Services

Field of Specialisation

  • Quantile Regression
  • Time Series Analysis

Journal articles

  1. I. Mudhombo & E. Ranganai. (2022). Robust Variable Selection and Regularization in Quantile Regression Based on Adaptive-LASSO and Adaptive E-NET. Computation. 10. 203. 10.3390/computation10110203.
  2. K. Chinhamu, R. Chifurira & E. Ranganai. (2022). Value-at-Risk Estimation of Precious Metal Returns using Long Memory GARCH Models with Heavy-Tailed Distribution. 89. 10.18576/jsap/110107.
  3. Maswanganyi, N., Sigauke, C. and Ranganai, E. Prediction of extreme conditional quantiles of electricity demand: An application using South African data, Energies (2021)., 14, 20 https: //doi.org/10.3390/en14206704
  4. E. Ranganai & C. Sigauke. Capturing Long-Range Dependence and Harmonic Phenomena in 24-Hour Solar Irradiance Forecasting: A Quantile Regression Robustification via Forecasts Combination Approach, DOI: 10.1109/ACCESS.2020.3024661, in IEEE Access, 2020: 8, 172204-172218.
  5. E. Ranganai & I. Mudhombo. Variable Selection and Regularization in Quantile Regression via Minimum Covariance Determinant based Weights, Entropy, 2021.
  6. T. J. Museba, E. Ranganai & G. Gianfrate. Customer perception of adoption and use of digital financial services and mobile money services in Uganda. Journal of Enterprising Communities: People and Places in the Global Economy (2021).
  7. K. Chinhamu, R. Chifurira & E. Ranganai,Value-at-risk estimation of precious metal returns using long memory GARCH models with heavy-tailed distributions, Journal of Statistics Applications & Probability (2021).
  8. E. Ranganai & L Matizirofa. An Analysis of Recent Stroke Cases in South Africa: Trend, seasonality and predictors. South African Medical Journal (2020).
  9. E. Afuecheta, C. E. Utazi, E. Ranganai & C. Nnanatu. An Application of Extreme Value Theory for Measuring Financial Risk in BRICS Economies. Annals of Data Science (2020).
  10. N. Maswanganyi, E. Ranganai & C. Sigauke. Long-term peak electricity demand forecasting in South Africa: A quantile regression averaging approach. AIMS Energy (2019).
  11. K Sivhugwana & E. Ranganai. Intelligent techniques, harmonically coupled and SARIMA models in forecasting solar radiation data: A hybridization approach. Journal of Energy in Southern Africa (2020).
  12. C. Sigauke, S. Kumar, N. Maswanganyi & E. Ranganai. Reliable Predictions of Peak Electricity Demand and Reliability of Power System Management. System Reliability Management: Solutions and Technologies (2018). Taylor & Francis.
  13. E. Ranganai & S. Nadarajah. A predictive leverage statistic for quantile regression with measurement errors. Communication in Statistics- Simulation and Computation, DOI:10.1080/03610918.2016.1204455, 2017: ISSN: 0361-0918 (Print) 1532-4141.
  14. E. Ranganai. Quality of fit measurement in regression quantiles: An elemental set method approach. Statistics & Probability Letters, 2016: 111, 18–25. ISSN: 0167-7152.
  15. E. Ranganai. On studentized residuals in the quantile regression framework. SpringerPlus, 5(1); DOI:10.1186/s40064-016-2898-6, 2016: 5, 1-11. ISSN: 2193-1801.
  16. E. Ranganai & S. B. Kubheka. Long Memory Mean and Volatility Models of Platinum and Palladium Price Return Series under Heavy Tailed Distributions. SpringerPlus, 5:2089; DOI: 10.1186/s40064-016-3768-y, 2016: 5, 1-20. ISSN: 2193-1801.
  17. E. Ranganai & M. B. Nzuza. A comparative Study of the Seasonal Autoregressive Integrated Moving Average (SARIMA) Models and Harmonically Coupled SARIMA Models in the Analysis and Forecasting of Seasonal Solar Radiation Data: A case study in Durban, South Africa. Journal of Energy in Southern Africa, 2015: 26, 125-137. ISSN 1021-447X online.
  18. E. Ranganai, J. O. van Vuuren & T. de Wet. Multiple Case High Leverage Diagnosis in Regression Quantiles. Communications in Statistics-Theory and Methods, DOI: 0.1080/03610926.2012.715225, 2014: 43, 3343–3370. ISSN: 0361-0926.
  19. N. Mswanganyi, C Sigauke & E. Ranganai. Peak Electricity Demand Forecasting Using Partially Linear Additive Quantile Regression Models, in the Proceedings of the 59th Annual Conference of SASA (2017), 1-8.

Professional positions, fellowships & awards

  • South African Statistical Association (SASA)
  • International Biometric Society (IBS)

Projects

  • Quantile Regression Diagnostics via Elemental Regression
  • Theory and Applications of Quantile Regression
  • Incidence of stroke under covariate measurement error.
  • Long Memory Time Series including GARCH and FIGARCH TYPE Models, Extremal Quantile Regression & Var Models.
  • Applications in renewable energy, precious metals.
  • Peak Electricity Demand Forecasting.
  • Characterization of renewable energy in Africa using Quantile Regression.
  • Multicollinearity Diagnosis in the Regression Quantile framework and penalty solution to multicollinearity.
  • Probabilistic Load and Renewable Energy (solar and wind) Forecasting.
  • Multilocation wind power nowcasting using wavelet Stochastic models and Neural Networks: A Nonlinear Denoising Via Wavelet Shrinkage Approach.
  • Some aspects of Quantile Regression Time Series Models.