The respective contributions of combustion noise, mechanical noise, noise attributed to the fuel injection system and other engine noise components must be identified in order to devise effective measures for reducing diesel engine noise. This paper describes a method for estimating the contributions of engine noise components using multiple regression analysis. With this method, the engine noise level is regarded as a criterion variable and the cylinder pressure level and several parameters are examined as explanatory variables. The accuracy of this method has been confirmed by comparing the results with experimental data. It was found that, by selecting adequate parameters as explanatory variables according to the type of fuel injection system used, the respective contributions of combustion noise, mechanical noise and fuel-injection noise estimated by multiple regression analysis agreed with the experimental data. By applying this multiple regression analysis method to actual engine development work, it is now possible to estimate these different engine noise components in a short time, and to reduce engine radiated noise more quickly.