This paper describes an identification of the best similar days parameters for artificial neural network (ANN) based short-term price forecasting. The work presented in this paper is an extended version of our previous works where we proposed the price prediction technique by using ANN, which is based on similar days method. According to similar days method, we select similar price days corresponding to forecast day based on Euclidean norm. The focus of the present paper is mainly on sensitivity analysis of similar days parameter while keeping the parameters same for ANN to forecast hourly electricity prices in the PJM electricity market. We simulated three cases by: (i) selecting two similar days parameters (load at t and price at t); (ii) selecting three similar days parameters (load at t, price at t and price at t - 1), and (iii) selecting five similar days parameters (load at t, load at t - 1, load at t + 1, price at t and price at t - 1). The next-24h price forecasts obtained from ANN based on similar days method confirm that the performance of the ANN model is better when five similar days parameters are selected, i.e., the accuracy of the method is enhanced by the addition of load at t - 1 and t + 1. The factors impacting the electricity price forecasting, including time factors, load factors, and historical price factors are well discussed. Mean absolute percentage error (MAPE) and forecast mean square error (FMSE) of reasonably small values were obtained for the PJM data, which has correlation coefficient of determination of (R2) of 0.7758 between load and electricity price.