Statistical Analysis Of The Framingham Heart Data

With the improvement of people’s living standard,cardiovascular disease(CVD)shows a rising trend year by year.And it has been becoming a major diseases that poses a threat to human health after the tumor.The prevention and control of myocardial infarction is imminent.Because there are many factors that can lead to myocardial infarction.Therefore,the research about how to find out main influencing factors from many factors and how to use these factors to judge if the individual has the risk of myocardial infarction is meaningful.Furthermore,the prevention and control of the disease is also meaningful.This paper used the logistic regression and decision tree to research Framingham cardiac data,and the goal of this paper is to establish the prediction model of myocardial infarction based on influencing factors.At present,the research on this respect is still relatively small in China.Decision tree and logistic regression are both very effective methods to solve the problem of "two classification".Among them,Logistic regression model is a kind of generalized linear regression model.It is often used in epidemiology,and it is mainly used to explore the risk factors of the disease.According to the risk factors,we can establish logistic regression model to predict the probability of a disease.Decision tree is a kind of simple but widely used classifier.The original data is divided into training set and testing set,then we can establish decision tree model on influencing factors of diseases.This paper preprocesses the Framingham heart data and makes descriptive statistical analysis on its first.Second,by establishing a logistic regression model,this paper analyzes the influencing factors that may lead to myocardial infarction and uses the spearman correlation analysis to select the independent variables of the model.Next,this paper calculates the relative weight of each variable.Through the relative weight value,the contribution of each factor to the myocardial infarction can be reflected.And I obtained the important influencing factors of myocardial infarction.Then,this paper divides original Framingham heart data into training set and test set.For the test set,I uses the decision tree model to judge if an individual will die in myocardial infarction in the 21 st time measurement.Finally,this paper compares the advantages and disadvantages of the two methods in the prediction that if an individual will have the myocardial infarction.