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

Time Paths of Weather-Induced Mood Effects on Stock Returns 66 The impulse response of local institutional investors is significant only on day 0. The average responses are positive on day 0 but are negative on days 1 to 9. The cumulative response is not significant. Although the responses for days 1 to 9 are not significant, the nonsignificant cumulative response implies the reversal. The weather effect for the local institutional investors’ trading behavior is temporary. The significant, temporary effects on the trading volume of local institutional investors are not strong enough to contribute to the return reversal, probably because the trading share of local institutional investors is small relative to that of local individual investors. Finally, the local proprietary investors’ net buying volume is not driven by the weather. Neither the cumulative response nor impulse responses are significant. If the trading of an investor group leads to weather effects in stock returns, the weather effects must also exist for the trading of that investor group as well (Goetzmann et al., 2015). From Tables 2 and 6, I find that weather effects exist for stock returns and the trading of local individual investors. There are no reversals of their responses to the weather. The effects are permanent. These results lead me to conclude that the weather- sensitive, local individual investors cause the weather effects in stock returns. 5.8 Impacts of Global Warming, Economic Conditions, and Investor Trading on Weather Effects The full sample is long, from 1992 to 2017. Over the 26 years, the weather conditions may have experienced some structural changes due to global warming. For the stock Table 6 Impulse Responses of Investor Groups ’ Net Buying Volume to One Standard Deviation Shock of Weather ’ s Fourth Principal Component Investor Group Impulse Response (Day) 0 1 2 3 4 Local Individual Investors 0.0100 0.0079 -8.38E-04 -8.90E-04 -5.14E-04 51.3762*** 61.5393*** 42.1131** 31.8966 23.0386 Local Institutional Investors -0.0263 0.0021 0.0070 0.0038 0.0017 36.0932* 20.3682 20.3966 18.2197 15.9952 Local Proprietary Investors 0.0011 -0.0069 0.0020 7.03E-04 3.76E-04 24.5564 34.5877 24.7837 15.7234 13.7305 Note: In each row, the numbers in the upper line are the average impulse responses for years 1992 to 2017, and the numbers in the lower lines are the χ 2 26 statistics for the joint hypotheses that the impulse responses are zero in all years from 1992 to 2017. *, **, and *** = significant at the