Abstract: If the coefficient matrix of the normal equation is ill-conditioned, the statistical cha-racter of adjustment parameter will become poor. To improve on the least squaresestimate, the ridge estimation and the stein estimation are proposed in this paper. Undercertain conditions, the mean square error of the ridge estimation and the stein estimationis smaller than the mean square error of the least square estimation. Different kinds ofmethods to obtain ridge coefficients and stein coefficients are introduced and somesuggestions on their use are given in this paper.