The implementation of the majority of energy maximising control strategies requires the knowledge of the wave excitation force experienced by the wave energy converter (WEC). In addition, many optimal numerical control strategies also require future knowledge, or a forecast, of future values of the excitation force. This paper examines both the excitation force estimation and forecasting problem for a heaving buoy wave energy device. In particular, a Kalman filter is used to estimate excitation force, where the wave force model is comprised of a set of oscillators at discrete frequencies. The forecasting algorithm consists of an autoregressive model, where the value of prefiltering, in terms of forecasting performance, is evaluated. The paper provides a level of sensitivity analysis of the estimation and forecasting performance to variations in sampling period, sea spectral shape factor and prediction horizon. Results demonstrate that the achievable performance of the estimator/forecaster is consistent with the broad requirements of numerical optimal WEC control strategies (Fusco and Ringwood (2012)), which depends on the characteristics of the radiation impulse response.