top of page


Eshetu Atenafu and Edit Gombay, Asymptotic Methods in Stochastics: Festschrift For Miklos Csorgo.American Mathematical Soc.2004 Page 301-313.

Screen Shot 2020-06-12 at 6.44.30 AM.png

This book contains articles arising from a conference in honour of mathematician-statistician Miklόs Csörgő on the occasion of his 80th birthday, held in Ottawa in July 2012. It comprises research papers and overview articles, which provide a substantial glimpse of the history and state-of-the-art of the field of asymptotic methods in probability and statistics, written by leading experts.

The volume consists of twenty articles on topics on limit theorems for self-normalized processes, planar processes, the central limit theorem and laws of large numbers, change-point problems, short and long range dependent time series, applied probability and stochastic processes, and the theory and methods of statistics. It also includes Csörgő’s list of publications during more than 50 years, since 1962.

Screen Shot 2020-06-12 at 6.46.54 AM.png

Khethan, V., Chan, H, Wang, L.,at al: Prognostic Factors in Cancer, 3E, M. K. Gospodarowics, Editor, J. Wiley & Sons, Inc., Hoboken, NJ, 39, 2006

M. K. Gospodarowicz, P. Hermanek, and D. E. Henson Attention to innovations in cancer treatment has tended to eclipse the importance of prognostic assessment. However, the recognition that prognostic factors often have a greater impact on outcome than available therapies and the proliferation of biochemical, molecular, and genetic markers have resulted in renewed interest in this field. The outcome in patients with cancer is determined by a combination of numerous factors. Presently, the most widely recognized are the extent of disease, histologic type of tumor, and treatment. It has been known for some time that additional factors also influence outcome. These include histologic grade, lymphatic or vascular invasion, mitotic index, performance status, symptoms, and most recently genetic and biochemical markers. It is the aim of this volume to compile those prognostic factors that have emerged as important determinants of outcome for tumors at various sites. This compilation represents the first phase of a more extensive process to integrate all prognostic factors in cancer to further enhance the prediction of outcome following treatment. Certain issues surround­ ing the assessment and reporting of prognostic factors are also considered. Importance of Prognostic Factors Prognostic factors in cancer often have an immense influence on outcome, while treatment often has a much weaker effect. For example, the influence of the presence of lymph node involvement on survival of patients with metastatic breast cancer is much greater than the effect of adjuvant treatment with tamoxifen in the same group of patients.

Lim S, Le LW, Hu P, Xing B, Greenwood CMT, Beyene J (2009). Integration of clinical, SNP, and microarray gene expression measurements in prediction of chronic fatigue syndrome. In Methods of Microarray Data Analysis(VI), McConnell P, Lim S, and Cuticchia AJ (Editors).  CreateSpace Publishing, Scotts Valley, United States 

Hu P, Le LW, Lim S, Xing B, Greenwood CMT, Beyene J (2009). Serum Diagnosis of Chronic Fatigue Syndrome Using Array-based Proteomics. In Methods of Microarray Data Analysis(VI), McConnell P, Lim S, and Cuticchia AJ (Editors).  CreateSpace Publishing, Scotts Valley, United States


In 2006, the Critical Assessment of Microarray Data Analysis (CAMDA) Conference was held at the Duke University Medical Center. The focus of the conference was methodologies used in analyzing microarray data based focused on Chronic Fatigue Syndrome. Chronic Fatigue Syndrome effects approximately 4 out of 1000 adults. Here we present eleven outstanding chapters containing manuscripts solicited from the conference.

Screen Shot 2020-07-04 at 2.11.38 PM.png

Competing Risks. A Practical Perspective.

For the data sets, please click here.


Author Melania Pintilie, Principal Biostatician

The need to understand, interpret and analyse competing risk data is key to many areas of science, particularly medical research. There is a real need for a book that presents an overview of methodology used in the interpretation and analysis of competing risks, with a focus on practical applications to medical problems, and incorporating modern techniques. This book fills that need by presenting the most up-to-date methodology, in a way that can be readily understood, and applied, by the practitioner.


Chapter: Model-free linkage analysis of a binary trait.


Authors: W Xu, SB Bull, L Mirea, CM Greenwood.

Screen Shot 2020-07-04 at 2.13.39 PM.png

Genetic linkage analysis aims to detect chromosomal regions containing genes that influence risk of specific inherited diseases. The presence of linkage is indicated when a disease or trait cosegregates through the families with genetic markers at a particular region of the genome. Two main types of genetic linkage analysis are in common use, namely model-based linkage analysis and model-free linkage analysis. In this chapter, we focus solely on the latter type and specifically on binary traits or phenotypes, such as the presence or absence of a specific disease. Model-free linkage analysis is based on allele-sharing, where patterns of genetic similarity among affected relatives are compared to chance expectations. Because the model-free methods do not require the specification of the inheritance parameters of a genetic model, they are preferred by many researchers at early stages in the study of a complex disease. We introduce the history of model-free linkage analysis in Subheading 1. Table 1 describes a standard model-free linkage analysis workflow. We describe three popular model-free linkage analysis methods, the nonparametric linkage (NPL) statistic, the affected sib-pair (ASP) likelihood ratio test, and a likelihood approach for pedigrees. The theory behind each linkage test is described in this section, together with a simple example of the relevant calculations. Table 4 provides a summary of popular genetic analysis software packages that implement model-free linkage models. In Subheading 2, we work through the methods on a rich example providing sample software code and output. Subheading 3 contains notes with additional details on various topics that may need further consideration during analysis.

Screen Shot 2020-07-04 at 2.16.05 PM.png

 Manual of Clinical Oncology

Chapter: Assessment of treatment outcome.


Authors: Manola J, Xu W, Giantonio B. 

Cancer studies frequently employ clinical endpoints for outcome reporting in order to estimate treatment effect sizes. Most often these outcome assessments use time‐to‐event measures in addition to tumour response, toxicity and quality of life (QOL). The Kaplan‐Meier method is often used to estimate the actuarial rate for time‐to‐event measures. Non‐stratified or stratified log‐rank tests are frequently applied assessing the treatment effect among groups. The Cox proportional hazards regression model is commonly used to estimate the hazard ratio between different treatments. Because cancer outcome is often confounded by multiple other outcomes (e.g. various causes of death), competing risks regression models are used to assess the treatment effect. In addition, intermediary endpoints, such as changes in tumour size, tumour‐related chemical markers and tumour metabolism may also assist in evaluating new treatments. Therefore, the ability to accurately and reliably assess the direct antitumour effect of investigational therapies is critical for the optimal conduct of clinical trials. The goal of this chapter is to summarize general principles of cancer outcome reporting and estimation of treatment effect, and response assessment.

bottom of page