The Forecasting Of The Most Respected Listed Company Based On Logistic Regression And Ridge Estimate

The contest of the most respected by investors listed companies has a great influence in China’ investors.The listed companies which were elected in this contest show the investors’ preferences effectively.Establishing appropriate model for this contest not only helps the listed companies to realize their social reputation,but also helps the investors to adjust the investment strategy.In fact,this forecast is a classification problem.When dealing with classification problems,we often use the logistic regression.The variables selected from financial indicators of listed companies often exist a serious multicollinearity problem.In this paper it is shown how ridge estimators can be used in Logistic regression to improve the parameter estimates and to diminish the error made by further predictions.In this example,we can find that the improved model can predict new observations more accurately.