Tse (1997) applied ARIMA model to the study of Hong Kong’s real estate prices. The model was essentially an approach to economic forecasting based on time series data. It helps to track the direction of changes in the real-estate prices. The study showed that ARIMA model produces forecasts that are likely to be more accurate than the forecasts produced by other approaches. The models had additionally tested to be excellent short-term forecasting models for a wide variety of time series because short-term factors were expected to alter slowly.
Along with the evidence from previous studies showed that ARIMA model had unlimited potential in forecasting. Junttila (2001) used an ARIMA model in the forms of recursive and rolling regressions with monthly data for the period 1978–1996 to examine the impact of structural breaks on the forecasts of Finnish inflation. An attempt to improve the forecasts from a univariate time series model by applying a rolling regressions technique and the structural break procedure, which together enable both small and large evolution in the coefficient of an ARIMA model for inflation. The advantage of the proposed model was that the nature, timing and size of the exogenous events affecting the rate of inflation were identified endogenously.
Besides, Chin and Fan (2005) calculated the price dynamics in the Singapore private housing market using the ARIMA modelling coupled with outlier detection and ARCH modelling techniques. In practice, univariate time series models, such as ARIMA models, are not only employed to identify the cyclical patterns and cyclical turning points of economic time series, but also to analyse the efficiency of the housing market. ARIMA models had proved to be very reliable for the short-term forecast of economic time series. It had even outperformed multivariate cointegration based models for short-term forecasts of some economic time series. In the modelling process, the researchers simultaneously considered the issue of the changing variance of residual series using outlier detection techniques and ARCH models. By doing so, the results will not only provide useful information about the property price variance, but also help to improve the accuracy of short-run forecast.
Tseng et al. (2001) developed a fuzzy ARIMA (FARIMA) model and applied it to forecasting the exchange rate of Taiwan Dollars (NTD) to US Dollars (USD). The resource data was the asking price of NTD/USD spot exchange rate between the bank and customers provided by The First Commercial Bank in Taiwan. It consists of 40 observations from 1 August 1996 to 16 September 1996. The first 30 observations were used to formulate the model and the next 10 observations were to evaluate the performance of the model. As a result, the proposed method not only can make good forecasts but also provides the decision makers with the best and worst-possible situations.