مقاله انگلیسی رایگان در مورد کشور های سرمایه گذار در تکنولوژی و بازده سهام – الزویر 2018

Recent literature has demonstrated that technology-investing countries have stock returns that are predictable using technology variables, such as patents granted and research and development; see, inter alia, Hirshleifer, Hsu, and Li (2013), Hsu and Huang (2010), and Hsu (2009). In this paper, we take this evidence as our starting point and as our motivation to ask a different but related question. We ask whether: (i) of all technology-investing countries (conditional on data availability), which factors other than technology predict returns?; and (ii) whether by using any of these recognized return predictors (exchange rate, inflation, U.S. stock excess returns, and oil price returns), can stock return forecasting models be proposed which could be useful for devising successful trading strategies? Our approaches to addressing these questions are fivefold. First, we select all countries for which historical time-series data (1981 to 2014) on both patents (number of patents granted to residents) and stock market price index are available. This search leads to a total of 77 countries, allowing us to rank countries (based on mean patents granted over the period 1981 to 2014) from the most technology-investing to the least. According to our ranking exercise, countries with the most number of patents granted are those that invest most in technologies and vice-versa. From this ranking, we also categorize countries into high and low patent investing countries, and high and low patent investing developed and emerging countries. This allows us to form eight different portfolios of countries apart from a portfolio consisting of all 77 countries. Second, because we have time-series data for each country, we choose from the broader literature on stock return predictability a list of predictors that have been successful in predicting stock returns in general. This search ends with the following choices. (1) We consider U.S. stock excess returns (proxied by the S&P500 price index returns in excess of the U.S. three-month Treasury bill rate) because recent studies find that U.S. stock excess returns are a powerful predictor of country returns (see Rapach et al. 2013; Nyberg and Ponka, 2016).‡ (2) We consider bilateral exchange rate returns (local currency vis-à-vis the U.S. dollar) based on the empirical evidence that exchange rate returns have predictive power (see Bartov and Bohnar, 1994; Phylaktis and Ravazzolo, 2005). (3) We also consider the inflation rate, which has traditionally been used as a predictor of stock returns (see Nelson, 1976; Fama and Schwert, 1977; Gultekin, 1983; Campbell and Vuolteenaho, 2004; Rapach et al., 2010; Gupta, and Modise, 2013). (4) The most popular predictor that dominates much of the recent literature on stock return predictability is the oil price; see Driesprong et al. (2008) and Phan et al. (2015).§ We, therefore, also use the oil price returns as a predictor.