臺大管理論叢 NTU Management Review VOL.30 NO.3

Time Paths of Weather-Induced Mood Effects on Stock Returns 68 in the time paths of stock return responses to the weather. The trading shares of individual and institutional investors vary in time. The (maximum, minimum) trading shares to these two investor groups are (75.01%, 45.83%) and (11.55%, 2.10%), respectively. Therefore, the time paths may behave differently with respect to the time-varying trading shares. I examine how time paths react to global warming, economic condition, and individual investors’ trading share. I regress the chi-square statistic for the cumulative response on global warming, economic condition, and trading share variables. The statistic measures the significance level of the responses to the weather (Doyle and Chen, 2009). I consider two choices for global warming variables and use them in two separate regressions. Following Novy-Marx (2014), the first variable is the global land-ocean temperature index. The index is the change in global surface temperature relative to 1951- 1980 average temperatures. It is retrieved from the National Aeronautics and Space Administration’s Goddard Institute for Space database. The second variable is a local measure of global warming, following Sandvik (2008) who uses CO 2 emissions metric tons per capita. I obtain the variable for Thailand from the countryeconomy.com database. The economic condition and trading share variables are the growth rate of real gross domestic product (GDP) and the annual trading share of local individual investors, respectively. The real GDP is from the Bank of Thailand database, and the annual trading shares are computed from the daily trading volumes reported by the SET. The regression results for the 1-, 10-, and 250-day cumulative responses are reported in Table 7. None of the coefficients are significant. Global warming, economic condition, and trading share have no influence on how the trading paths behaved over the sample. The fact that global warming does not affect the time paths may be because the effects are a long-term global trend of the temperature, while the result in Table 5 indicates that there is no temperature effect in the SET. The insignificant relationship of economic conditions with the time path may be due to inertia with respect to changing wealth (Brunnermeier and Nagel, 2008). Finally, the trading share of local individual investors cannot explain the time paths probably because of the continual dominance of individual investors over the sample period. 5.9 Weather-Induced Investor Inattention From Table 6, local individual investors are weather-sensitive, and they determine the time path of stock returns. Hong and Yu (2009) and Schmittmann et al. (2015) find that for the U.S. and German markets, respectively, individual investors trade less on days with