Although statistical models for changes in the temperature of carcass surfaces have been developed to obtain scientific evidence for the HACCP plan, little research has been conducted into interval estimation. The aim of this paper was to develop a statistical model for surface temperature changes of cattle carcasses during cold storage and to compare methods for interval estimation. The model parameters were estimated from the observed surface temperature changes of 53 cattle during cold storage by the method of least squares. Then, the confidence intervals of parameters, the confidence intervals and the prediction intervals of carcass surface temperatures were calculated using four methods: first order Taylor series approximation, the inverse of the Fisher information matrix, parametric bootstrap and jackknife. Results did not vary significantly among the four methods. Therefore, it is assumed that these four methods can be used to calculate the interval estimation of the carcass surface temperature.