Univariate non-causal models - Box-Jinkens ARIMA Models and traditional exponential smoothing methods, are used in this paper to study the tourist arrivals in Macau. Statistical computer package SAS (Statistical Analysis System) is used to perform the forecasting. Five univariate time-series forecasting models are tried to fit the data, and their forecasting accuracy are compared. These models are Naïve I, simple exponential smoothing, Holt's linear trend exponential smoothing, and Winter's exponential smoothing, and SARIMA model. The number of monthly tourist arrivals in Macau is divided into two data sets: The initialization data set (Jan. 1990 to Dec. 2000) is used to fit the models; The resulting models are evaluated on the hold-out data set (from Jan. 2001 to Dec. 2002) in terms of the forecasting ability of the models. A one-year ahead forecasting and a two-year ahead forecasting are compared. Among the five models, Winter's exponential smoothing is the best one to fit the data, ARIMA(0, 1, 1)(0, 1, 1)₁₂ model is the optimal one to produce the forecast. The ranking of the forecasting ability of the five forecasting methods is ARIMA (0, 1, 1) (0,1,1)₁₂, Winter's exponential smoothing, Holt's exponential smoothing, simple exponential smoothing and Naïve I if MAPE is used as the measure of forecasting accuracy. MAPE has improved 7% if we use ARIMA (0,1,1)(0,1,1)₁₂ model instead of Naïve I method to forecast. One-year ahead forecast is significantly more accurate than two-year-ahead forecast when using ARIMA model and Winter's model. SAS is an appropriate and powerful tool to analyze the time series data, and to model and forecast the time series data. Winter's forecast could be combined with ARIMA forecast to produce a better forecast and the intervention model could be added to handle the impact of the special events. Since the tourist arrivals in Macau is a rather complex time series data, and ARIMA model has been shown to be well suited to handle complex time series and is optimal in forecasting such data. Therefore, we suggest that the government of Macau should use the Box-Jenkins ARIMA technique to further analyze and forecast the tourist arrivals in Macau.