Logistic Regression Model in Predicting Diabetes of Pima Indians Female Population

Tea, Khai Yun (2019) Logistic Regression Model in Predicting Diabetes of Pima Indians Female Population. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Tea Khai Yuin_FULL TEXT.pdf Restricted to Registered users only Download (2MB)

Abstract

In this Final Year Project, we investigate the diabetes problem happened on Pima Indian female population by using Logistic Regression. The report consists of five chapters. In Chapter 1, we introduce the statistics of diabetes which happened in the global, the types of diabetes, and the percentage of patients who were death from diabetes disease. In section 1 of Chapter 1, we present some definitions and terminologies related to the topic of diabetes. In Section 2 of Chapter 1, we present the objectives of doing this analysis on diabetes disease. In Chapter 2, we referred to other people’s researches which related to the diabetes topic and do some conclusion on it. In Section 1 of Chapter 2, we present some learning lessons on the method of logistic regression from the references book. In Section 2 of Chapter 2, we present some literature review which related to the diabetes topic from other’s project. In Chapter 3, we study the methodology of logistic regression. In Section 1 of Chapter 3, we study the logistic regression method, the assumption of this method, and the tools we use to test the data set which is IBM SPSS 25.0. In Section 2 of Chapter 3, we present the data collection and requirement. In Chapter 4, we present the main results of the whole data set. In Section 1 of Chapter 4, we checked the assumption of the logistic regression before running the data. In Section 2 of Chapter 4, the main result was generated by SPSS software. We tried to remove some of the predictor variables which are not significant contribute to the model and check does the model still significant after removing the certain predictor variable. In Chapter 5 which is the last chapter, we present the conclusion of the results and also some future work and suggestions for further improvement.