The analyses on the transfiguration of urban land-use pattern have been carried out by using the land-use transition matrix. However, the effect of the error in the transition matrix is still not well understood. This is the problem discussed here. We attempt to construct a method for evaluating the estimation error caused by the variance in the transition matrix. As a numerical example, the proposed method is applied to the actual land-use data, and we evaluate the confidence intervals of the estimated values.