State estimation in power engineering is used as a tool to find the unknown parameter values from the hypothesized model by utilizing the specified information available about the system. Due to random noises that are added from different sources, the exact value of the state vector cannot be found. This study is an effort to describe the simultaneous and individual confidence intervals for the state parameters in view of the heteroscedastic structure of the error terms. The performance of the constructed intervals in terms of coverage probability has been evaluated by using the Monte Carlo simulation study. The results of the study demonstrate that it is an effective method for practical implementation in the state estimation a power system. © Medwell Journals, 2017.