Current optimization of power forecasting parameters in park integrated energy systems focuses on minimizing the forecasting error of either the source or the load power, without considering the comprehensive impact of forecasting results on operation objectives. This oversight leads to lower comprehensive economic benefits when the forecasting parameters are applied in practice. To address this issue, this paper proposes a joint optimization method for power forecasting parameters in park integrated energy systems, oriented towards operational target. Firstly, based on a two-stage day-ahead and intraday operational model, a joint optimization objective function for forecasting parameters is constructed, integrating total economic operational indices from both day-ahead and intraday stages. Subsequently, based on linear regression equation, this paper establishes a joint optimization model for source-load power forecasting parameters with an embedded two-stage operational strategy. Simultaneously, a warm-start scheme for the forecasting model is designed to accelerate the solving speed of the joint optimization problem of forecasting parameters. Finally, leveraging commercial and residential energy consumption data from Anchorage, USA, a simulation model is constructed to validate the proposed method. The validation results indicate that the proposed method can effectively reduce the operating costs of urban park integrated energy systems, decrease the curtailment of wind and solar energy, and increase the operational revenue of energy storage.