Two-level nested simulation methods have been recently applied for the analysis of simulation experiments under parameter uncertainty. On the outer level of the nested run, we generate (n) observations of the parameters, while on the inner level; we fix the parameter on its corresponding value and generate (m) observations using a simulation model. In this paper, we focus on the output analysis of two-level stochastic simulation experiments for the case where the observations of the inner level are independent, showing how the variance of the simulated observations can be decomposed in the sum of parametric and stochastic components. Furthermore, we derive central limit theorems that allow us to compute asymptotic confidence intervals to assess the accuracy of the simulation-based estimators for the point forecast and the variance components. Theoretical results are validated through experiments using a forecasting model for sporadic demand, where we have obtained analytical expressions for the point forecast and the variance components.