目的 – 本研究旨在探討國際股指尤其是發達國家股指對發展中國家股指的有效波動組合。未來更多，提供有效的國際波動組合作為不確定環境下的行業指南。設計/方法/方法– 本研究應用了五種不同的時間序列模型；指數加權移動平均線 (EWMA)、自回歸綜合移動平均線 (ARIMA)、廣義自回歸條件異方差 (GARCH)、霍爾特線性趨勢 (Holt 趨勢) 和一般加權移動平均線 (GWMA)；通過嘗試與道瓊斯工業平均指數 (DJIA)、日本日經 225 指數、恆生指數 (HSI) 和台灣證券交易所 (TAIEX) 相關的國際股票指數組合來預測泰國證券交易所 (SET) 指數的每日收盤價。通過在線提供獲取，收集二手數據作為樣本期。然後，應用誤差測量以測試所提出模型的準確性和效率。發現 – 結果表明，預測SET指數的最佳時間序列預測模型是廣義自回歸條件異方差（GARCH）和自回歸綜合移動平均（ARIMA）模型，可以提供最小的MSE和MAPE。此外，主要的國際股票波動效應是道瓊斯工業平均指數（DJIA）。研究限制– 在泰國進行的胸部和工業解釋和建議是有限的，不能為通用框架解決。此外，可以在不同的構建模型中應用相互關係和數學方法，以提高產量並獲得更好的產業方向。實際影響– 對於金融和/或經濟行業在 SET 指數預測方面，GARCH 模型與 DJIA 國際股票的波動效應相關聯，推動了未來 SET 每日收盤價預測的更高準確性。原創性/價值 – 現有的研究並沒有關注時間序列模型的比較，非常有限的研究對比較模型進行了檢驗，並且只關注時間序列模型或顯著變量中的一個影響來獲得模型實現。 Purpose – This study aims to investigate the effective volatility combination of international stock index especially on developed countries on stock index in developing country. Future more, provide the valid international volatility combination to be an industrial guideline under uncertain environment.Design/methodology/approach – This study applies five different time series models; Exponential Weighted Moving Average (EWMA), Autoregressive Integrated Moving Average (ARIMA), Generalized Auto-Regressive Conditional Heteroscedastic (GARCH), Holt’s Linear Trend (Holt trend), and generally weighted moving average (GWMA); to forecast daily closing price on Stock Exchange of Thailand (SET) index by attempt international stock index combination associated with Dow Jones Industrial Average (DJIA), Japan's Nikkei 225, Hang Seng Index (HSI) together with Taiwan Stock Exchange (TAIEX). Secondary data is collected as the sample period by online provide acquirement. Then, error measurements are applied in order to test an accuracy and efficiency of proposed model. Findings – The result showed that the best time series forecasting model to predict SET index is Generalized Auto-Regressive Conditional Heteroscedastic (GARCH) and Inferior with Autoregressive Integrated Moving Average (ARIMA) model which can provide the smallest MSE and MAPE. Moreover, the main international stock volatility effect is Dow Jones Industrial Average (DJIA).Research Limitation – The thoracal and industrial explanation and suggestion conducted were limited in Thailand and could not be addressed for generalized framework. Moreover, interrelationship and mathematical methodologies could be applied in different constructed model to improve output and obtain the better industrial direction.Practical Implications – For financial and/or economic industry in term of SET index forecasting, GARCH model associate with the volatility effect from DJIA international stock is driven higher accuracy for future SET daily closing price prediction. Originality/Value – The existing studies were not focus on comparison of time series models and very limited studies have examined for comparison model and focus on one influence either time series model or significant variable to acquire the model implementation.
