@article{Chaengkham_Wianwiwat_2021, title={Stock market index prediction using machine learning: evidence from leading Southeast Asian countries}, volume={39}, url={https://so05.tci-thaijo.org/index.php/TER/article/view/251892}, abstractNote={<p>Financial investment in stock markets in emerging economies has played an important role in wealth management. Thus, this study aimed to present the application of the Support Vector Machine (SVM) with using the Toda-Yamamoto causality test to select economic indicators for predicting the movement of one month ahead of the stock market index in four leading Southeast Asian (ASEAN) nations: Indonesia, Malaysia, Singapore, and Thailand. Monthly data were sampled, ranging from January 2002 to December 2019. The linear kernel SVM provided useable results with accuracy ranging from 58.14 % to 65.12 %, which performed better than the sigmoid kernel SVM. According to the efficient-market hypothesis (EMH), the stock markets of Singapore and Malaysia were the most efficient among the four stock markets, whereas investors could strategically utilise this SVM algorithm to gain more returns from stock markets in Thailand and Indonesia.</p>}, number={2}, journal={Thailand and The World Economy}, author={Chaengkham, Supakorn and Wianwiwat, Suthin}, year={2021}, month={Jun.}, pages={56–64} }