Industry network on returns predictability of Returns

• Main Authors: CHANG, SHU-YU, 張書宇
• Other Authors: Lee, Han-Hsing
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/u347a4

碩士 === 國立交通大學 === 財務金融研究所 === 103 === Previous studies have documented how information diffusion and economic shocks can change with the position in the inter-industry network. Our empirical result herein provides evidence that both supplier portfolio returns and customer portfolio returns can cross-predict firm-level stock returns, and both of them are positively related to future stock returns. We also find that lagged own-industry, lagged supplier portfolio returns, and lagged customer portfolio returns of the central industries whose centrality are greater than 0.165 have better predictability at firm-level stock returns than those of non-central industries. Our result suggests that central industries are associated more strongly with linked firms than non-central industries. We also investigate the self-financing trading strategies under various industry setting and market conditions, showing that a supplier strategy in the bear market yields a maximum return of 5.8% per annum. In addition, the performance of the supplier strategy increases considerably in the stable market, but the performance of the supplier strategy decreases and the performance of the customer strategy increase in the volatile market.