The purpose of this econometric analysis is to model the concentration of particle number emissions from a hybrid diesel-electric bus in terms of operating characteristics. Important aspects of this study are that particle number concentrations are modeled instead of particle mass, and the emissions are recorded using on-board instrumentation in real-world driving conditions. The operating characteristics included in the final models are two engine parameters: fuel rate and engine speed and two vehicle parameters: velocity and acceleration. The emissions data possess properties that frequently cause problems in regression analysis: nonstationarity, multicollinearity, heteroscedasticity, and autocorrelation. Methods for overcoming and/or minimizing the effects of these properties are implemented. The Newey-West autocorrelation consistent covariance estimator is implemented using ordinary least squares (OLS) to produce a model that accounts for heteroscedasticity and autocorrelation, without requiring assumptions to be made about the structure of the model disturbances. A first-order autoregressive process is used with feasible generalized least squares as a comparative model. Both models have similar coefficients, fit and predictive capability. However the models are specific in scope to typical freeway driving conditions. In future studies, the researchers anticipate applying econometric analysis to model particle number emissions among different routes, bus technologies, aftertreatments, and atmospheric conditions.