Curriculum Vitae

PERSONAL DATA

RESEARCH

1. Areas of main research interest

• Data mining and text mining
• Data quality
• New robust procedures
• Statistical computing
• Bioinformatics

2. Main Contributions to the Theory of Robustness

At the beginning of my research career in 1986 I was mainly interested in the general theory of quantitative robustness, which was dominated by the concepts of influence function and breakdown point. I thought that these measures of quantitative robustness gave an incomplete assessment of the degree of robustness of an estimate. I also thought that a much better assessment can be made using the concept of maximum asymptotic bias (maxbias) introduced, for the simple location model by Huber (1964). Huber quickly abandoned this approach because it, in his opinion, led to a "rather uneventful theory". The maxbias approach was "dormant" for nearly 20 years until I showed in my Ph.D. dissertation that maxbias functions and minimax bias estimates can be derived for other models including scale, regression and orthogonal regression. Together with collaborators (Victor J. Yohai and R. Doug Martin) we showed that the minimax bias theory can be extended to linear regression and found minimax bias robust regression estimates. It is safe to say that today the maxbias approach is considered the most important theoretical tool in quantitative robustness. My current work on this area is aimed at showing that the maxbias approach can also give rise to useful statistical procedures, via the construction of bias bounds and globally robust inference. I wish to create and popularize inferential procedures which take into account the possible effect of contamination bias. Together with collaborators (V.J. Yohai, J. Adrover, J.R. Berrendero and M. Salibian) we developed a new globally robust confidence intervals of minimax length and initiated the new robustness theory which we call "global robust inference". The aim of global robust inference is to construct robust confidence intervals, p-values and tests that take into account not only the uncertainty due to "normal" data variability but also the bias effect of "abnormal" noise and poor data quality.

3. Current Research Interest and Directions

I am currently mostly interested in data mining and statistical computing and pursue a fruitful collaboration with researchers in Computer Science including Raymond Ng, Alan Wagner and Laks Lakshmanan. We are co-supervising several graduate students who are working on different data mining problems. A paper that resulted from this collaboration won the "Best Paper Award" in the KDD 2001 Conference. We are now working on the scaling of robust algorithms using parallel computing, on text mining and compression of large relational databases. Another main achievement from this collaboration is the three year funding (2001/2003) obtained from MITACS for the project "Toward Interactive Data Mining". Our MITACS project had two main industrial partners: Insightful Corporation (the producer of Splus and I Miner) and IBM through a collaboration with the iCAPTURE Centre at the St Paul's Hospital.

I am also interested in the study and modelling of data quality in the context of large, high dimensional datasets. Together with my former student, Fatemah Alqallaf, we proposed a new, flexible model to represent contamination in multivariate data. This model provides a mathematical formulation for a phenomenon that we call "outlier propagation". I think that outlier propagation is a serious statistical problem and that our model may become an important tool in the context of the robust analysis of high dimensional datasets.

I am interested in statistical problems in genomics. To pursue this interest I established a fruitful collaboration with Dr. McManus and his research laboratory. Professor McManus is the Co-Director of iCAPTURE and the Director of the Cardiovascular Research Laboratory and the Cardiovascular Registry, Department of Pathology and Laboratory Medicine. The exciting project entitled "Better Biomarkers of Acute and Chronic Allograft Rejection" recently obtained $9.1 million of funding over three years. This project will generate very interesting statistical problems as well as financial support for several Statistics graduate students and one or two postdoctoral fellows.

4. New Statistical Procedures

-Estimates: Victor Yohai and I introduced the class of robust -estimates. We defined these estimates for the case of linear regression. Now -estimates have also been defined for multivariate location, orthogonal regression, principal component, etc. -estimates can attain breakdown point of 1/2 and arbitrary high efficiency at the "target" model.

Orthogonal Regression M-Estimates: I defined these estimates and studied their robustness properties.

Image Enhancement: Jean Meloche and I proposed a generally applicable, non-parametric method to enhance and restore binary images.

Robust (Fast) Bootstrap: Matias Salibian and I developed a new bootstrap method to quickly estimate the variability of some computational intensive robust regression estimates. Our robust bootstrap is not only several orders of magnitude faster than classical bootstrap but also gives more stable and reliable variance estimates in the presence of outliers and data contamination.

Linear Grouping Algorithm (LGA): together with collaborators (Stefan Van Aelst, Steven Wang and Rong Zhu) we proposed a new approach to clustering where items following similar linear relationship are grouped together.

CLUES: together with collaborators (Steven Wang and Weiliang Qiu) we proposed a new approach to clustering based in k-near neighbors averaging. Iteratively, each point is replaced by its local average and the procedure is continued to convergence.

Pairwise Correlation Matrices: together with collaborators (Ricardo Maronna and Fatemah Alqallaf ) we proposed several robust correlation/covariance estimates based on pairwise operations. Two of these methods (pairwise QC and pairwise GK) are now available in the Splus 6.1 robustness library.

STUDENTS AND TRAINEES

Post-Doctoral Fellows

Ph.D. Students

Master Students

PUBLICATIONS

1. Tomal, J.H. Welch, W.J. and Zamar, R.H. (2016) "Exploiting Multiple Descriptor Sets in QSAR Studies" Journal of Chemical Information and Modeling. 56 (3), pp 501 - 509
2. Zhang, H., Leung, A. and Zamar, R.H. (2016) "Robust sparse hierarchical clustering. Submitted.
3. Kondo, Y., Salibian-Barrera, M. and Zamar, R.H. (2016). "RSKC: An R Package for a Robust and Sparse K-means Clustering Algorithm". To appear in the Journal of Statistical Computing
4. Leung, A., Zhang, H. and Zamar, R.H. (2016) "Robust regression estimation and inference in the presence of cellwise and casewise contamination". Computational Statistics and Data Analysis vol 99, 1-11.
5. Tomal, H. W.J., Welch, W and Zamar, R.H. (2015). "Ensembling classification models based on phalanxes of variables with applications in drug discovery". The Annals of Applied Statistics 2015, Vol. 9, No. 1, 69-93.
6. Agostinelli, C., Leung, A., J. Yohai, V.J., Zamar, R.H. (2015). Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination. Test, 24, 441-461
7. Agostinelli, C., Leung, A., J. Yohai, V.J., Zamar, R.H. (2015). Rejoinder on: Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination. Test, 24, 441-461
8. Zhang, H. and Zamar, R.H. (2013). Discussion of "How to Find an Appropriate Clustering for Mixed Type Variables with Application to Socio Economic Stratication" by Christian Hennig and Tim F. Liao. Journal of the Royal Statistical Society: Series C (Applied Statistics), 62(3), 362-363.
9. Zhang, H. and Zamar, R.H. (2013). "Least Angle regression for model selection", Advanced Review in WIREs Comput Stat 2013. doi: 10.1002/wics.1288. 2013
10. Van Aelst, S, Willems, G. and Zamar, R.H.(2013). "Robust and efficient estimation of the residual scale in linear regression", Journal of Multivariate Analysis 116, 278--296
11. Pena, D., Viladomat, J. and Zamar, R.H. "Nearest-neighbours median cluster algorithm". Statistical Analysis and Data Mining 349-362. (2012)
12. Cohen Freue, G.V., Ortiz-Molina, H. and Zamar, R.H. (2013). "A natural robustication of the classical instrumental variables estimator". Biometrics, 2013 . Sep; 69(3):641-50. doi: 10.1111/biom.12043. Epub 2013 Jul 19.
13. Danilov, M., Yohai, V.J. and Zamar, R.H. (2012). "Robust estimation of multivariate location and scatter in the presence of missing data", Journal of the American Statistical Association, vol. 107(499), pages 1178-1186.
14. Boente, G., Ruiz, M. and Zamar, R.H. (2012). "Bandwidth choice for robust nonparametric scale function estimation". Computational Statistics and Data Analysis vol. 56, 6, 1594-1608.
15. Yan, G., Welch, W.J. and Zamar, R.H. (2010) Model-Based linear clustering, Can. J. of Statistics, Vol. 38, No. 2, 2010, Pages 1-22. paper
16. Pena, D., Zamar, R.H. and Yan, G. (2008). "Bayesian likelihood robustness in linear models". To appear in J. of Statistical Planning and Inference. ;  paper   
17. Garcia-Escudero, L.A., Gordaliza, A., San Martin, R., Van Aelst, S., and Zamar, R.H. (2009). " Robust Linear Clustering", Journal of the Royal Statistical Society B, 71 (2), 1-18.   paper   
18. Ghement, R.I., Ruiz, M. and Zamar, R.H. (2008)."Robust Estimation of Error Scale in Nonparametric Regression Models". J. of Statist. Planning and Inference, 138, 3200-3216.   paper   
19. Alqallaf, F., Van Aelst, S.V., Yohai, V.J. and Zamar, R.H. (2009) "Propagation of Outliers in Multivariate Data". The Ann. of Statist., 37, 311-331.   paper   
20. Salibian-Barrera, M., Willems, G. and Zamar, R.H. (2008). "The fast-Tau-estimator for regression". Computational and Graphical Statistics, Vol. 17, 659-682. paper   
21. Gabriela V. Cohen Freue, Zsuzsanna Hollander, Enqing Shen, Ruben H. Zamar, Robert Balshaw, Andreas Scherer, Bruce McManus, Paul Keown, W. Robert McMaster, and Raymond T. Ng (2007)."MDQC: A New Quality Assessment Method for Microarrays Based on Quality Control Reports". Bioinformatics 23 (2007) 3162-3169. paper
22. Khan, J.A., Van Aelst, S., and Zamar, R.H. (2007), "Robust Linear Model Selection Based on Least Angle Regression," Journal of the American Statistical Association, Vol. 102, No. 480, pp.1289-1299. paper & software
23. Podder, M., Welch, W.J. Zamar, R.H. and Tebbutt, S.J. (2006). " Dynamic Variable Selection in SNP Genotype Autocalling from APEX Microarray Data", BMC Bioinformatics 2006, 7:521 paper
24. Willems, G., Joe, H. and Zamar, R.H. " Diagnosing multivariate outliers detected by robust estimators". paper   
25. Cohen-Freue, G., Ortiz-Molina, H. and Zamar, R.H. "A Natural Robustification of the Ordinary Instrumental Variables Estimator". Submitted to Biometrics "A robust instrumental variables estimator". Submitted 2006 paper   
26. Khan, J., Van Aelst, S. and Zamar, R.H. (2007). "Building a robust linear model with forward selection and stepwise procedures." Computational Statistics & Data Analysis, Volume 52, Issue 1, 239-248. paper   
27. Danilov, M., Ng, R., Yohai, V.J. and Zamar, R.H. "Robust Regression by Cone Reweighting". In preparation.
28. Chilson, J., Ng R., Wagner, A., and Zamar, R.,"Parallel Computation of High Dimensional Robuts Correlation and Covariance Matrices". Algorithmica, Special Issue on Coarse Grain Computation, Springer New York, Vol. 45, No. 3, pp. 403-431, 2006. paper   
29. Zhang, R.Y., Lakshmanan L.V.S. and Zamar, R.H. (2004). "Extracting XML Data from HTML Repositories", SIGKDD Newsletter, Volume 6, Issue 2, pg 5-13. paper   
30. Wang X., Qiu W. and Zamar, R.H. (2007). "CLUES: A non parametric clustering method based on local shrinking". Computational Statistics & Data Analysis, Volume 52, Issue 1, 286-298 paper   
31. Van Aelst, S., Wang, X., Zamar, R.H. and Zhu, R. (2006). " Linear Grouping Using Orthogonal regression," Computational Statistics and Data Analysis, 50, 1287-1312." paper    package talk
32. Chilson, J., Ng, R., Wagner, A. and Zamar, R. (2004). "Parallel computation of high dimensional robust correlation and covariance matrices", Proceedings of the ACM SIGKDD, 533-538. paper   
33. Berrendero, J.R. and Zamar, R.H. (2004). "A note on the uniform asymptotic normality of location M-estimates". To appear in Metrika.
34. Yohai, V. and Zamar, R.H. (2004). "Robust non-parametric inference for the median". Ann. of Statist., Vol. 32, No. 5, 1841--1857.
35. Salibian-Barrera, M. and Zamar, R.H. (2004). "Uniform asymptotic for robust location estimates when scale is unknown". Ann. of Statist. Vol. 32, No. 4, 1434--1447.
36. Adrover J. and Zamar, R.H (2004). "Bias robustness of three median-based regression estimates", J. of Statist. Planning and Inference. 203-227.
paper   
37. Adrover, J., Salibian, M. and Zamar, R.H. (2004). "Globally robust inference for location and the simple linear regression model". J. of Statist. Planning and Inference, 119, 353-375. paper   
38. Adrover, J., Berrendero, J.R., Salibian, M. and Zamar, R.H. "Globally robust inference". (2002). Estadistica, 54, 127-161.
39. Maronna, Ricardo and Zamar, R.H. (2002). "Robust multivariate estimates for high dimensional data sets". Technometrics, 44 (4) 307-317.
40. Aqallaf, F.A., Konis, K..P., Martin, R.D. and Zamar, R.H. (2002). "Scalable robust covariance and correlation estimates for data mining" . Proceedings of the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data, Mining, Edmonton, Alberta 14-23. paper   
41. Svartz, M., Yohai, V.J., and Zamar, R.H. (2002). "Optimal bias robust M--estimates of regression", submitted for the proceedings of the International Conference on L1 and Related Methods, Neuchatel, Switzerland, 191- 200
42. Park, J. and Zamar, R.H. (2001). "The detection and testing of multiple outliers in linear regression". Bulletin of the International Statistical Institute. 53rd Contributed Papers -- Tome LIX, Book 1 -- Seoul, Korea. 473-474.
43. Yohai, V.J. and Zamar, R.H. (2001). "A review of recent robust inference: an approach based on maxbias". Bulletin of the International Statistical Institute. 53rd Session Proceedings -- Seoul, Korea. 505-508.
44. Knorr, E., Ng, R. and Zamar, R.H. (2001). "Robust space transformation for distance based operations". Proceedings of the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, 126-135.
45. Salibian, M. and Zamar, R.H. (2002). "Bootstrapping robust estimates of regression". The Ann. of Statist. 30, 556-582 .
46. Berrendero, J.R. and Zamar, R.H. (2001). "The maxbias curve of robust regression estimates". The Ann. of Statist. 29, No. 1, 224-251
47. Fraiman, R., Yohai, V.J. and Zamar, R.H. (2001). "Optimal robust m-estimates of location". The Ann. of Statist. 29, No. 1, 194 - 223.
48. Heckman, N. and Zamar, R.H. (2000). "Comparing the shape of regression functions". Biometrika. 87, 1, pp. 135-144.
paper   
49. Berrendero, J.R. and Zamar, R.H. (1999). "Global robustness of location and dispersion estimates". Statist. and Probability Letters, 44, 63-72.
50. Ferretti, N, Kelmansky, D., Yohai, V.J. and Zamar, R.H. (1999). "A class of locally and globally robust regression estimates". J. Amer. Statist. Ass. 94, 174-188
51. Berrendero, J., Mazzi, S., Romo, J. and Zamar, R.H. (1998). "On the Explosion Rate of Maxbias Functions". appear Can. J. of Statist., 26, 333-351
52. Justel, A., Pena, D. and Zamar, R.H. (1997). "A multivariate Kolmogorov-Smirnov test of goodness of fit". Stat. and Probab. Letters, 35, 251-59
53. Pena, D. and Zamar, R.H. (1997). "A simple diagnostic tool for local prior sensitivity". Statist. and Probab. Letters, 36, 205-212.
54. Yohai, V.J. and Zamar, R.H. (1997). "Optimal locally robust M-estimates of regression". Journal of Statist. Planning and Inference, Vol. 66, 2, 309-323.
55. Pena, D. and Zamar, R.H. "Bayesian Robustness: An Asymptotic Approach". (1996). In Robust Statistics, Data Analysis, and Computer Intensive Methods (H. Rieder, Ed.). Springer Lecture Notes in Statistics #109, 361- 374.
56. Li, B. and Zamar, R.H. (1996). "M-estimates of regression when the scale is unknown and the error distribution is possibly asymmetric: a minimax result". Can. J. of Statist., Vol. 24, 2, 193-206.
57. Zamar, Ruben H. "Estimacion Robusta". With discussion. (1994). Estadistica Espanola. Vol. 36, No. 137, 317-387
58. Meloche, J. and Zamar, R.H. (1994). "Binary-image restoration". Can. J. of Statist., Vol 22, 4, 335-355.
59. Maroma, R., Yohai, V.J. and Zamar, R.H. (1993). "Bias-robust regression estimation: A partial survey". In New Directions in Statistical Data Analysis and Robustness. S. Morgentaler, E. Ronchetti and W.A. Stahel (Ed.), Birkhauser, Basilea, 157-176.
60. Yohai, V.J. and Zamar, R.H. (1993). "A minmax-bias property of the least Alpha-quantile estimates". The Annals of Statistics, 21, No. 4, 1824-1842.
61. Martin, R.D. and Zamar, R.H. (1993). "Bias robust estimates of scale". The Annals of Statistics, 21, No. 2, 991-1017 .
62. Martin, R.D. and Zamar, R.H. (1993). "Efficiency constrained bias robust estimation of location". The Annals of Statistics, 21, No. 1, 338-354,.
63. Zamar, R.H. "Bias robust estimation in orthogonal regression". The Annals of Statistics, 20, 1875-1888, (1992).
64. Le, N. and Zamar, R.H. (1992). "A global test for effects in 2k factorial designs without replicates". Journal of Statistical Computation and Simulation, 41, 41-54,.
65. Yohai, V.J., Stahel, W. and Zamar, R.H. (1991). "A procedure for robust estimation and inference in linear regression". In: Directions in Robust Statistics and Diagnostics, Part II, W. Stahel and S. Weisberg (Eds), Springer-Verlag, 365-374, .
66. Yohai, V.J. and Zamar, R.H. (1991). "Bounded influence estimation in the errors-in-variables model". In: Statistical Analysis of Measurement Error Models and Applications, Philip J. Brown and Wayne A. Fuller (Editors), Contemporary Mathematics, 112, 243-248,.
67. Li, B. and Zamar, R.H. (1991). "Min-max asymptotic variance when scale is unknown". Statistics and Probability Letters, 11, 139-145,.
68. Zamar, R.H. (1990). "Robustness against unexpected dependence in the location model". Statistics and Probability Letters, 9, 367-374,.
69. Martin, R.D., Yohai, V.J. and Zamar, R.H. (1989). "Min-max bias robust regression". Annals of Statistics, 17, 1608-1630, .
70. Martin, R.D. and Zamar, R.H. (1989). "Asymptotically min-max bias-robust M-estimates of scale for positive random variables". Journal American Statistical Association 84, 494-501,.
71. Zamar, R.H. "Robust estimation in the errors in variables model" (1989). Biometrika, 76, 149-60.
72. Yohai, V.J. and Zamar, R.H. (1988). "High breakdown-point estimates of regression by means of the minimization of an efficient scale". Journal American Statistical Association, 83, 406-413..
    
    

    ---

Recent Invited Talks

• Invited speaker at the Workshop in Honor of Victor Yohai, Madrid, Spain, October, 2006.
• Invited speaker at ICORS 2006, Lisbon, Portugal, July, 2006.
• Invited speaker, 25th European Meeting of Statisticians , Oslo July 24-28 2005. Invited Session "IS16 - NEW DIRECTIONS IN ROBUST STATISTICS".
• "Robust Methods and Data Mining". Workshop on Data Mining Methodology and Applications at The Fields Institute, Toronto, October 28-30, 2004
• Keynote speaker (opening talk) at the Workshop on Robustness for High-Dimensional Data, Sponsored by the European Science Foundation, Vorau, Austria, May 5-8, 2004.
• Keynote speaker (opening talk) at the CASCON 2003 Workshop on Categorical Data Analysis. Toronto, November, 2003.
• "Pair-wise approach to Scatter Matrix Estimation", International Conference on Statisitics, Combinatorics and Related Areas, Maine, USA, October 2003.
• "Robust Model Selection", ICORS 2003, Antwerp, July 13-18, 2003.
• "A new model for data contamination", SCC, Invited session on Data Mining, Halifax, June 8-12, 2003.
• "Robust Bootstrap". 24th European Meeting of Statisticians. Prague, August 19-23, 2002.
• "Optimally robust regression M-estimates", Fourth International Conference on Statistical Data Analysis Based on L1 Norm and Related Methods, Newchatel, Switzerland, August 4-9, 2002.

  "A Review of Recent Robust Inference: An Approach Based on Maxbias", 53rd International Statisitical Institute, Seoul, Korea, August 2001.

• "Three Median based regression estimates", Minerva Conference on Robust Statistics, B. Aires, Argentina, 2001.
• "Non-Parametric and Robust Confidence Intervals", International Conference on Robust Statistics (ICORS 2001), Vorau, Austria, July 2001.
• "Robust covariance for very large datasets", Joint Statistical Meetings, Atlanta, USA, August 2001
• "Comparing the shape of regression functions", V ICSA International Conference, Hong Kong, China, August 2001.

Recent Research Support

EDITORIAL ACTIVITIES

• Associate Editor, Annals of Statistics 2004-2007
• Associate Editor, Test since 1998.
• Associate Editor, Journal of the American Statistical Association, 1991-93.
• Reviewer for the Mathematical Reviews since 2000.
• Member of KDD-2003 Program Committee, Washington, DC, U.S.A.
• Co-Editor of the Proceedings of ICORS 2003, Antwerp, Belgium.
• Co-Editor of the Proceedings of ICORS 2001, Vorau, Austria.

CONFERENCE ORGANIZATION

• Co-organizer (with Stefan Van Aelst, Ghent University, Belgium) of the Focused Research Group on Robust Analysis of Large Data Sets. Banff International Research Station. June 5 - 19, 2004.
• Member of Scientific Committee, Symposium on Data Mining, IMS - mini meeting, University of Ghent, May 2004.
• Member of Scientific Committee of ICORS 2003, June 2003, Antwerp, Belgium.
• Co-organizer (with Arnold Stromberg) of the "Workshop on Robust Computational Procedures 2002" . This workshop, held at PIMS, received funding from NSF and PIMS.
• Co-organizer (with Ursula Gather and Chris Field) of ICORS 2002, June 2002, Vancouver, Canada. This conference received funding from PIMS and the Minerva Research Foundation.
  
  

AWARDS

• "Best Paper Award" International Conference on Knowledge Discovery and Data Mining 2001 (KDD2001) San Francisco, jointly with Ed Knorr and Raymond Ng.
• "2001/02 Science Research Achievement Awards", Faculty of Science, UBC.