Staff profile

Qualifications

• B.Sc.(Hons), Ph.D. (La Trobe).

Teaching areas

• Applied statistics at second year, sample surveys and multivariate analysis at third year and regression analysis including dimension reduction at 4th year.

Supervision

• Recent Honours Students:
  2008. Connie Li Wai Suen (Power series expansions for eigenvalue and eigenvector estimators under contamination).
  2008. Michael Malloy (The influence function: A useful diagnostic tool for robustness).
  2008. Amanda Shaker (SIR and SAVE for discrete binary classification).
  2007. Nathan Bock (An influence diagnostic for single-index model ordinary least squares dimension reduction).
  2006. Callum McLean (Sliced Inverse Regression).
  2005. Jodie Smith (The Influence Function).
• Recent Masters by Coursework Students:
  2006. Lindsey Kevan (Influence function considerations for principal component analysis).
• Recent Postgraduate Diploma Students:
  2005. Mitra Jazayeri (An overview of smoothing techniques and its application to high dimensional data).
• Current Postgraduate Research Students (as a Co-supervisor)
  Michael Malloy (Meta-regression).
  David Farchione (Confidence intervals utilizing prior information).
• 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)
  Jodie Smith (Robustification of principal Hessian directions).
  Connie Li Wai Suen (A formal robustness study of meta-analysis).
• Current Honours Students
  Alexandra Garnham (Principal Hessian directions utilizing response transformations).
  Magdalene Mahadavan (The lack of residual checking in published research).

Professional involvement

• I am a member of the Statistical Society of Australia Inc. (SSAI) and 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 publications and presentations

• Prendergast, L. A. (2008), Trimming influential observations for improved single-index model estimated sufficient summary plots, Presentation at the Australian Statistical Conference, Melbourne, June 30 – July 3.

• 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.