Modeling and Analysis Method of National Fitness Big Data for Basketball Projects Based on a Multivariate Statistical Model

How to start from the fitness needs of people and effectively improve precision supply public services everyone is an important issue that be solved first at current stage. This requires us proceed reality, conduct accurate research, find a method can match problem. In this paper, taking basketball projects in national as example, by introducing proposition about development small events, corresponding big data modeling analysis methods are studied. The research objectives involved paper based on relevant parameters multivariate statistical model. First, article introduces calculation principle multiple linear regression We introduce concept variance inflation factor carry out variable. order illustrate application effect kind three different technologies, including immune selection optimization algorithm, particle swarm Elman neural network, predict analyze (VIF) project. results show network exhibits certain advantages terms computing convergence time. And, number steps increases, superiority more obvious. As far prediction performance concerned, square correlation coefficient algorithm largest sum squares residuals smallest, showing superior performance.