The researchers developed a method called Restricted Ridge Regression to deal with multicollinearity in multiple linear regression analysis. This method adds a constant bias to the calculation, resulting in more accurate parameter estimates. By using prior information about the parameters, the Restricted Ridge Regression method outperformed the traditional Ordinary Least Square method in terms of accuracy. In a practical application, the parameter estimates obtained through Restricted Ridge Regression had the smallest mean square error compared to those obtained through Ordinary Least Square.