Forecasting is an estimation of the systematic process which is most likely to occur in the future based on past informations. Wavelet is one of forcasting method without parameter, which is used in signal analysis, data compression, and time series analysis. On the other hand ARIMA is the most general class of models for forecasting a time series, which can be stationarized by transformations, such as differencing and logging. This research present the forecasting of gold price in Indonesia using wavelet and ARIMA. The results show that wavelet gives the value of Mean Square Error (MSE) which is smaller than the ARIMA. Therefore wavelet is considered quite well in the analysis of time series data.