Dr Garth Tarr

Lecturer in Statistics and Data Science

F07 - Carslaw Building
The University of Sydney

Telephone +61 2 9351 5776

Website Google Scholar
Twitter
Research Gate
Curriculum vitae Curriculum vitae

Biographical details

Garth Tarr is a statistician and data scientist with expertiese in feature selection in complex data and predictive modelling. He has strong industry ties, particularly with meat processing and pastoral industries, and he is the statistical and methological advisor on the Meat Standards Australia Beef Pathways Committee. He joined the University of Sydney as a lecturer in 2017 after a previous appointment at the University of Newcastle (2015-17) and postdoc at the Australian National university (2014-15).

Research interests

Garth Tarr is a member of the Statistics research group and the Statistical Bioinformatics and Biometrics research group. His research interests include robust statistics, model selection, data visualisation, meat science, econometric modelling, educational research as well as biostatistics and biometrics. He develops new statistical methods and software to help scientists and researchers make more informed decisions about the statistical models they build. Garth's research is driven by applications in modelling complex agricultural data and biostatistics.

Teaching and supervision

Research students

Short courses

Model selection with R.

  • Australian National University, Canberra, Australia, 10-11 April 2017 (29 participants)
  • University of Western Australia, Perth, Australia, 23-24 November 2015 (30 participants).

Fast algorithms and modern visualisations for feature selection. International Society for Clinical
Biostatistics and Australian Statistical Conference, Melbourne, Australia, 26 August 2018.

Analytics for industry.

  • Meat and Livestock Australia, Brisbane, Australia, 17-19 January and 1-2 February 2018.
  • Teys Australia, Brisbane, Australia, 24-26 July and 22-23 August 2017 (11 participants).
  • Meat and Livestock Australia, Brisbane, Australia, 15-19 May 2017 (8 participants).
  • Australian Country Choice, Brisbane, Australia, 7-11 November 2016 (11 participants).

Data visualisation, interactive data analysis and statistical programming. BioInfoSummer, Sydney, Australia, 11 December 2015 (90 participants).

Model selection with R.
Australian National University, Canberra, Australia, 10-11 April 2017 (29 participants)
University of Western Australia, Perth, Australia, 23-24 November 2015 (30 participants). Fast algorithms and modern visualisations for feature selection. International Society for Clinical
Biostatistics and Australian Statistical Conference, Melbourne, Australia, 26 August 2018.
Analytics for industry.
Meat and Livestock Australia, Brisbane, Australia, 17-19 January and 1-2 February 2018. Teys Australia, Brisbane, Australia, 24-26 July and 22-23 August 2017 (11 participants). Meat and Livestock Australia, Brisbane, Australia, 15-19 May 2017 (8 participants).
Australian Country Choice, Brisbane, Australia, 7-11 November 2016 (11 participants).
Data visualisation, interactive data analysis and statistical programming. BioInfoSummer, Sydney, Australia, 11 December 2015 (90 participants).
Model selection with R.
Australian National University, Canberra, Australia, 10-11 April 2017 (29 participants)
University of Western Australia, Perth, Australia, 23-24 November 2015 (30 participants). Fast algorithms and modern visualisations for feature selection. International Society for Clinical
Biostatistics and Australian Statistical Conference, Melbourne, Australia, 26 August 2018.
Analytics for industry.
Meat and Livestock Australia, Brisbane, Australia, 17-19 January and 1-2 February 2018. Teys Australia, Brisbane, Australia, 24-26 July and 22-23 August 2017 (11 participants). Meat and Livestock Australia, Brisbane, Australia, 15-19 May 2017 (8 participants).
Australian Country Choice, Brisbane, Australia, 7-11 November 2016 (11 participants).
Data visualisation, interactive data analysis and statistical programming. BioInfoSummer, Sydney, Australia, 11 December 2015 (90 participants).

Timetable

GM_Tarr

Current projects

  • Are consumers willing to pay for eating quality? (Summer project, 2018)
  • Outlier identification and classification with functional data (Summer project, 2018)
  • Improved model averaging through better model weights
  • Finite sample performance of robust location estimators
  • Optimal robust location estimation in a bounded interval
  • Statistical issues in evaluating the eating quality of meat

Associations

  • Statistical Society of Australia
  • Australian Mathematical Society
  • American Statistical Association
  • International Biometrics Society
  • Priority Research Centre for Computer-Assisted Research Mathematics and its Applications, University of Newcastle (CARMA)

Awards and honours

  • Vice-Chancellor’s Award for Teaching Excellence and Contribution to Student Learning, University of Newcastle. With Renner I, 2016.
  • EJG Pitman Prize for the most outstanding talk presented by a young statistician at the Australian Statistical Conference, 2012.
  • First prize for the best oral presentation by a student at the Australian Young Statisticians Conference, 2013.
  • Statistical Society of Australia Golden Jubilee Travel Grant, 2013.
  • Vice-Chancellor’s Award for Systems that Achieve Collective Excellence in Teaching and Learning, University of Sydney. With Morison K and Morr J, 2011.

Software

International links

United States

(Cornell University) Research collaboration with Ines Wilms

Selected grants

2017

  • Advanced measurement technologies for globally competitive Australian meat value chains; Tarr G; Meat and Livestock Australia Ltd/Research and Development Grant.

Selected publications

Download citations: PDF RTF Endnote

Journals

  • Konarska, M., Kuchida, K., Tarr, G., Polkinghorne, R. (2017). Relationships between marbling measures across principal muscles. Meat Science, 123, 67-78. [More Information]
  • Tarr, G., Mueller, S., Weber, N. (2016). Robust estimation of precision matrices under cellwise contamination. Computational Statistics and Data Analysis, 93, 404-420. [More Information]
  • Dancer, D., Morrison, K., Tarr, G. (2015). Measuring the effects of peer learning on students' academic achievement in first-year business statistics. Studies in Higher Education, 40(10), 1808-1828. [More Information]
  • Mueller, S., Tarr, G. (2015). Much more than U-statistics: A symposium to celebrate Neville C. Weber. Statistical Society of Australia. Newsletter, 151, 10-12.
  • Tarr, G. (2015). Quantile based estimation of scale and dependence. Bulletin of the Australian Mathematical Society, 92(1), 173-175. [More Information]
  • Tarr, G., Weber, N., Mueller, S. (2015). The difference of symmetric quantiles under long range dependence. Statistics and Probability Letters, 98, 144-150. [More Information]
  • Tarr, G., Mueller, S., Weber, N. (2012). A robust scale estimator based on pairwise means. Journal of Nonparametric Statistics, 24(1), 187-199. [More Information]
  • Tarr, G. (2012). Small sample performance of quantile regression confidence intervals. Journal of Statistical Computation and Simulation, 82(1), 81-94. [More Information]

Conferences

  • Muir, M., Drury, H., Tarr, G., White, F., Morrison, K. (2015). A Start to Fostering Knowledge, Skills and Attitudes to Ensure Inclusive Teaching Environments. 12th annual conference of the International Society for the Scholarship of Teaching and Learning. International Society for the Scholarship of Teaching and Learning.
  • Muir, M., Tarr, G., Drury, H., White, F., Morrison, K. (2014). Knowing your students: An approach to encourage pedagogical creativity and change. International Society for the Scholarship of Teaching and Learning (ISSOTL), Not applicable: Not applicable.
  • Tarr, G. (2013). Efficient and robust scale estimation. 2011 International Conference on Robust Statistics, Valladolid, Spain: Universidad de Valladolid.
  • Tarr, G. (2013). Robust covariance estimation via quantiles of pairwise means and applications. International Conference on Robust Statistics.
  • Tarr, G. (2013). Robust covariance estimation with Pn. Australian Statistical Conference, Collingwood, Australia: CSIRO Publishing.
  • Tarr, G. (2013). Robust scale and autocovariance estimation. European Meeting of Statisticians.
  • Tarr, G. (2013). Robust scale estimation with extensions. Young Statisticians Conference, Melbourne, Australia: University of Melbourne.
  • Tarr, G. (2009). Quantile regression confidence intervals: theoretical exposition and empirical findings. 2009 Young Statisticians Conference, Collingwood, Australia: CSIRO Publishing.

Other

  • Tarr, G. (2013), Robust estimation of scale and covariance with Pn and its application to PCA. University of NSW, School of Mathematics and Statistics Seminar, Sydney, Australia.
  • Tarr, G. (2013), Robust estimation of scale and covariance with Pn and its application to precision matrix estimation. University of Sydney, School of Mathematics and Statistics Seminar, Sydney, Australia.

2017

  • Konarska, M., Kuchida, K., Tarr, G., Polkinghorne, R. (2017). Relationships between marbling measures across principal muscles. Meat Science, 123, 67-78. [More Information]

2016

  • Tarr, G., Mueller, S., Weber, N. (2016). Robust estimation of precision matrices under cellwise contamination. Computational Statistics and Data Analysis, 93, 404-420. [More Information]

2015

  • Muir, M., Drury, H., Tarr, G., White, F., Morrison, K. (2015). A Start to Fostering Knowledge, Skills and Attitudes to Ensure Inclusive Teaching Environments. 12th annual conference of the International Society for the Scholarship of Teaching and Learning. International Society for the Scholarship of Teaching and Learning.
  • Dancer, D., Morrison, K., Tarr, G. (2015). Measuring the effects of peer learning on students' academic achievement in first-year business statistics. Studies in Higher Education, 40(10), 1808-1828. [More Information]
  • Mueller, S., Tarr, G. (2015). Much more than U-statistics: A symposium to celebrate Neville C. Weber. Statistical Society of Australia. Newsletter, 151, 10-12.
  • Tarr, G. (2015). Quantile based estimation of scale and dependence. Bulletin of the Australian Mathematical Society, 92(1), 173-175. [More Information]
  • Tarr, G., Weber, N., Mueller, S. (2015). The difference of symmetric quantiles under long range dependence. Statistics and Probability Letters, 98, 144-150. [More Information]

2014

  • Muir, M., Tarr, G., Drury, H., White, F., Morrison, K. (2014). Knowing your students: An approach to encourage pedagogical creativity and change. International Society for the Scholarship of Teaching and Learning (ISSOTL), Not applicable: Not applicable.

2013

  • Tarr, G. (2013). Efficient and robust scale estimation. 2011 International Conference on Robust Statistics, Valladolid, Spain: Universidad de Valladolid.
  • Tarr, G. (2013). Robust covariance estimation via quantiles of pairwise means and applications. International Conference on Robust Statistics.
  • Tarr, G. (2013). Robust covariance estimation with Pn. Australian Statistical Conference, Collingwood, Australia: CSIRO Publishing.
  • Tarr, G. (2013), Robust estimation of scale and covariance with Pn and its application to PCA. University of NSW, School of Mathematics and Statistics Seminar, Sydney, Australia.
  • Tarr, G. (2013), Robust estimation of scale and covariance with Pn and its application to precision matrix estimation. University of Sydney, School of Mathematics and Statistics Seminar, Sydney, Australia.
  • Tarr, G. (2013). Robust scale and autocovariance estimation. European Meeting of Statisticians.
  • Tarr, G. (2013). Robust scale estimation with extensions. Young Statisticians Conference, Melbourne, Australia: University of Melbourne.

2012

  • Tarr, G., Mueller, S., Weber, N. (2012). A robust scale estimator based on pairwise means. Journal of Nonparametric Statistics, 24(1), 187-199. [More Information]
  • Tarr, G. (2012). Small sample performance of quantile regression confidence intervals. Journal of Statistical Computation and Simulation, 82(1), 81-94. [More Information]

2009

  • Tarr, G. (2009). Quantile regression confidence intervals: theoretical exposition and empirical findings. 2009 Young Statisticians Conference, Collingwood, Australia: CSIRO Publishing.

To update your profile click here. For support on your academic profile contact .