Abstarct: When multicollinearity among the columns of the design matrix of the linear regression exists, using the least squares method, is usually too weak to obtain efficient estimates of regression coefficients. It has been proved that the variance of least squares estimates may significantly increase and the length of parameter’s vector get too big. In the presence of multicollineatity in this context, one way to overcome the problems, is using the biased estimators. Several methods such as ridge regression, Liu and etc proposed for obtaining biased estimates of regression coefficients. One of these methods is using the modified Jackknife Liu estimator that is the combination of the Liu and Jackknife Liu estimators. However, this estimator is biased but has lower variance and also has a better performance comparing to the least squares estimator. In this thesis, we have studied the ridge regression model. Using the results of Akdeniz Duran and Akdeniz (2012) and Li and Yang (2012), we construct a new estimator, namely the modified jackknife Liu estimator. Given the complexity of the proposed estimator, the Monte Carlo numerical methods are provided to illustarte the bias and risk of the namely proposed estimator.