Staff profile

Qualifications

• BSc(Hons), PhD(La Trobe).

Teaching areas

• Data-based critical thinking at first year, linear models at third year and regression analysis at fourth year.

Supervision

• Current Honours Students:
  • Jacinda Barnard, Using area under the curve to analyse repeated measures data
• Recent Honours Students:
  • Alan Healey, 2010. Some considerations on the Lasso.
  • Alexandra Garnham, 2009. Principal Hessians directions and response transformations.
  • Magdalene Mahadavan, 2009. Do researchers consider residual checking to be an integral part of multiple linear regression analysis?
  • Connie Li Wai Suen, 2008. Power series expansions for eigenvalue and eigenvector estimators under contamination.
  • Michael Malloy, 2008. The influence function: A useful diagnostic tool for robustness.
  • Amanda Shaker, 2008. SIR and SAVE for discrete binary classification.
  • Nathan Bock, 2007. An influence diagnostic for single-index model ordinary least squares dimension reduction.
  • Callum McLean, 2006. Sliced Inverse Regression.
  • Jodie Smith, 2005. The Influence Function.
• Recent Masters by Coursework Students:
  • Lindsey Kevan, 2006. Influence function considerations for principal component analysis.
• Recent Postgraduate Diploma Students:
  • Mitra Jazayeri, 2005. An overview of smoothing techniques and its application to high dimensional data.
• Current Postgraduate Research Students (as a Co-supervisor)
  • Michael Malloy. Meta-regression.
• Recent Postgraduate Research Students (as a Co-supervisor)
  • Mitra Jazayeri. Approximating cross-validation results for SIR and SAVE preceded by principal component dimension reduction. Thesis Passed November 2008.
• Current Postgraduate Research Students (as a Principal Supervisor)
  • Alexandra Garnham. Improving estimated sufficient summary plots using response transformations.
  • Jodie Smith. Robustification of principal Hessian directions.
  • Connie Li Wai Suen. A formal robustness study of meta-analysis.
  • Amanda Shaker. Iterative application of inverse regression methods.
  • Alan Healy. Improving Estimated Sufficient Summary Plots.

Professional involvement

• Member of the Statistical Society of Australia Inc (SSAI)
• Member of the Institute of Mathematical Statistics (IMS).

Research interests

• I am interested in robustness properties of statistical estimators, in particular with respect to dimension reduction methods and meta analysis. I am also interested in the visualization of high dimensional data after dimension reduction. I am currently involved in clinical trial research involving weight loss studies.

Recent peer-reviewed publications

• Shaker, A. J. and Prendergast, L. A. (2011), Iterative application of dimension reduction methods, Electronic Journal of Statistics, Vol. 5, pages 1471-1494.
• Sumithran, P. and Prendergast, L. A. and Delbridge, E. and Purcell, K. and Shulkes, A. and Kriketos, A. and Proietto, J. (2011), Long-term persistence of hormonal adaptations to weight loss, New England Journal of Medicine, Vol. 365, pages 1597-1604.
• Malloy, M., Prendergast, L. A and Staudte, R. G. (2011), Comparison of Methods for Fixed Effect Meta-Regression of Standardized Differences of Means, Electronic Journal of Statistics, Vol. 5. Pages 83-101.
• Prendergast, L. A, and Li Wai Suen, C. (2011), A new and practical influence measure for subsets of covariance matrix sample principal components with applications to high dimensional datasets, Computational Statistics and Data Analysis, Vol. 55, pages 752-764.
• Prendergast, L. A. and Smith, J. A. (2010), Influence Functions for Dimension Reduction Methods: An Example Influence Study of Principal Hessian Direction Analysis, Scandinavian Journal of Statistics, Vol. 37, pages 454-467.
• Delbridge, E. A., Prendergast, L. A., Pritchard, J. E. and Proietto, J.  (2009), One-year weight maintenance after significant weight loss in healthy overweight and obese patients: Does diet composition matter?, American Journal of Clinical Nutrition, Vol. 90, pages 1203-1214.
• Prendergast, L. A. (2008), Trimming influential observations for improved single-index model estimated sufficient summary plots, Computational Statistics and Data Analysis, Vol. 52, pages 5319-5327.
• Prendergast, L. A. (2008), A note on sensitivity of principal component subspaces and the efficient detection of influential observations in high dimensions, Electronic Journal of Statistics, Vol. 2, pages 454-467.
• Prendergast, L. A. (2007), Implications of influence function analysis for sliced inverse regression and sliced average variance estimation, Biometrika, Vol. 94, pages 585-601.
• Prendergast, L. A. and Smith, J. A. (2007), Sensitivity of principal Hessian analysis, Electronic Journal of Statistics, Vol. 1, pages 253-267.