It has become a popular research topic how data on Internet queries are used to make reliable predictions about changes in the marketplace. In this study we analyze the relationship between the online search volume of IPO or initial public offering stocks and their post-IPO stock returns. We obtain the online search data sets of the NAVER, one of the largest online search services in Korea, and the information on IPO stocks and their post-IPO stock returns at the Korea Exchange. We investigate the daily online search volume data of 87 companies that went public in the year of 2016. After analyzing the relations of the abnormal returns and online search volumes using the event study methods, we find that the lower the amount of online search for stocks before IPO, the higher the stock returns after IPO both in short and long-term. One standard deviation increase in search volume decreases one-day return by 3 basis points, two-day return by 6 basis points, and 4-day return by 10 basis points. These economically significant results become stronger if we control for benchmark returns, firm size, and book-to-market ratio. This finding suggests that IPO stocks with low investors’ attention based on the Internet search volume may be undervalued.

如何使用Internet查询数据来对市场变化做出可靠的预测已成为一个流行的研究主题。在这项研究中，我们分析了IPO或首次公开发行股票的在线搜索量与其IPO后股票收益之间的关系。我们获得了韩国最大的在线搜索服务机构之一NAVER的在线搜索数据集，以及在大韩交易所的IPO股票及其首次公开发行后的股票回报信息。我们调查了2016年上市的87家公司的每日在线搜索量数据。使用事件研究方法分析了异常收益与在线搜索量的关系后，发现股票的在线搜索量越低首次公开发行之前，无论短期还是长期，首次公开募股之后的股票收益都较高。搜索量增加一标准差，则一日收益降低3个基点，两日收益降低6个基点，而4天回报降低10个基点。如果我们控制基准回报，公司规模和市帐率，那么这些具有经济意义的结果将变得更加强大。这一发现表明，基于互联网搜索量的，投资者关注度较低的IPO股票可能被低估了。