Associate Editor of Computational Statistics, Journal of Energy Markets, and Surveys in Mathematics and its Applications, his research focuses on risk management and forecasting in the power markets and computational statistics as applied to finance and insurance. His other interests include stochastic modeling, time series, heavy tailed distributions, and computer simulations of highly volatile phenomena. In this paper we study simple time series models and assess their forecasting performance. In particular we calibrate ARMA and ARMAX (where the exogenous variable is the system load) processes. Models are tested on a time series of California power market system prices and loads from the period proceeding and including the market crash.