http://rulrepository.ru.ac.bd/handle/123456789/121 2026-04-07T23:11:01Z http://rulrepository.ru.ac.bd/handle/123456789/1105 The Role of High Leverage Points in Regression Diagnostics Khan, Md. Ashraful Islam In fitting a linear regression model by the least squares’ method, leverage values play a very important role. They often fo1m the basis of regression diagnostics as measures of influential observations in the explanatory variables. Much work has been done on the detection of high leverage values and a good number of diagnostic measures are now available in the literature. But neither of these methods is effective in the identification of high leverage points when multiple high leverage points are present in the data. In our study we proposed a new method for the identification of multiple high leverage points. The usefulness of this newly proposed method is studied under a variety of leverage structures through Monte Carlo simulation experiments. We also investigated the performance of the newly proposed method as a remedy to multi collinearity problem caused by the presence of multiple high leverage points. This Thesis is Submitted to the Department of Statistics, University of Rajshahi, Rajshahi, Bangladesh for The Degree of Master of Philosophy (MPhil) 2003-01-01T00:00:00Z http://rulrepository.ru.ac.bd/handle/123456789/1056 Comparison of Performances of Machine Learning Techniques in Healthcare Data Maniruzzaman, Md. Due to the increasing prevalence of diabetes and cancer, it is an urgent need to develop automated system that helps to detect disease using one of the modern technologies. Nowadays, Machine Learning (ML)-based methods have become very popular as an automatically model building techniques. Despite of the rapid development of theories for computational intelligence, application of ML-based classifiers to diabetes and cancer diagnosis remains a challenging issue. Still these ML-based classifiers did not give a satisfactory accuracy and therfore cannot correctly classify healthcare data like diabetes and cancer patients. Because most of the diabetes and cancer dataset are complex in nature and contains missing values, unusual observations, multi-collinearity problems and so on. In most of the existing research, the researcher did not use feature selection (FS) techniques to identify the risk factors of cancer and diabetes disease. They applied limited classifiers to classify and predict the diabetes and cancer status but they did not tune the hyper parameter of the classifiers, as a result, their accuracy and AUC were low. Thus, an attempt has been made in this study to increase the accuracy of the classifiers in diabetes and cancer data by considering the above factors in ML-based algorithm. The main objective of this study is to comparison the performances of ML-based methods in healthcare data and suggests the best model with better performance compared to the models published in the existing research.----- This Thesis is Submitted to the Department of Statistics , University of Rajshahi, Rajshahi, Bangladesh for The Degree of Master of Philosophy (MPhil) 2021-01-01T00:00:00Z http://rulrepository.ru.ac.bd/handle/123456789/993 Robust Diagnostic Deletion Techniques in Linear and Logistic Regression Nurunnabi, Abdul Awal Md. Identification of unexpected observations is a topic of great attention in modem regression analysis. At the beginning statisticians differ but now they recognize robust regression and regression diagnostics are two complementary remedies to study unusual observations. We use both of them for identifying irregular observations at a time. We find out the group deletion diagnostic methods that show better performance for identifying influential observations in linear regression. These are based on robust regression and/or relevant diagnostic methods so that these are free from huge computational tasks and reliable in presence of masking and/or swamping because of prior suspect-group identification. We find a technique that performs well in case of large number and high-dimensional data sets. We have done a classification task of unusual observations in linear regression according to their nature of consequences on the analysis, and model building process. At the same time the method performs well for identifying influential observations. This method may be a good addition to the existing graphical literature. We have seen that our proposed procedures in linear regression are also effective to the logistic regression after some modification and development to the existing identification techniques in linear regression. Our further contribution is to propose two new identification techniques for influential observations in logistic regression. The new methods show efficient performance for the proper identification of unusual observations and thereby provide less misclassification error in the response variable for the binomial logistic regression. Summarizing all the above issues we can say that we have made contribution in three areas: identification of influential observations in linear regression, classification of unusual observations in linear regression, and identification of unusual observations in logistic regression. This Thesis is Submitted to the Department of Statistics, University of Rajshahi, Rajshahi, Bangladesh for The Degree of Master of Philosophy (MPhil) 2008-01-01T00:00:00Z http://rulrepository.ru.ac.bd/handle/123456789/983 A Study of Some Non-linear Stochastic Models on Renewable Resources Management: Application to Forestry Hossain, Md. Mowazzem Management is a vibrant and multi phase dynamic process, just like social, scientific and engineering processes, that involves a wide spectrum of activities such as planning, coordination, communication, policy framing, decision making and their implementation. During the last three decades, the management of natural resources in general and that of renewable resources, in particular, has invited the attention of a large segment of researchers in various field [1-SJ. In order to maintain the ecological balance, as well as, meet the economic needs, the forest can play a vital role. Our government also taking special initiative to can-y forward a forestation program throughout the country. Also, to pollution free environment a forestation has no alternative. In this direction government organizations came forward to utilize the renewable resources for economic prosperity of the nation. This Thesis is Submitted to the Department of Statistics, University of Rajshahi, Rajshahi, Bangladesh for The Degree of Master of Philosophy (MPhil) 2010-01-01T00:00:00Z