some CONTRIBUTIONS TO PREDICTIVE STATISTICAL MODELLING WITH APPLICATIONS TO ENDOMETRIAL CANCER

Abstract

Cancer is a major chronic disease in the twenty first century. Worldwide, it is the leading cause of death after cardiovascular diseases and has emerged as a major global public health problem. Endometrial cancer (EC) is a type of gynaecological cancer commonly found in adult women. It is the sixth most common type of cancer(s) among women worldwide. However, fourth most common type in India. It is noticed that 4.8% of all cancers diagnosed in women is endometrial cancer. Statistical modelling can be a vital tool for understanding the epidemiology of EC, which can contribute towards the controlling of the disease and suggesting better treatment to the patients. In this thesis, we have made an attempt to develop and use statistical models to estimate survival probability, and determine significant factors associated with the progression of EC. Furthermore, a Classification and regression tree (CART) predictive model is proposed newline