Summary: The maximum likelihood method is used to estimate the shape parameter in Pareto distribution under type-I doubly censored sample. Due to the complexity of the likelihood equation, the explicit expression of parameter estimation cannot be obtained. But, it is proved that the maximum likelihood estimation is uniquely existent. Because EM algorithm is an effective way to deal with defective data, here, the algorithm is used to estimate the parameter. The iterative formula and asymptotic variance of the shape parameter are obtained by EM algorithm, the approximate confidence interval of the shape parameter is obtained according to the central limit theorem. The stochastic simulation results show that the EM estimation of the shape parameter converges to its maximum likelihood estimation. An example is given for point estimation and interval estimation of the parameter under different samples.