4 Empirical Results
4.2 The Effect of Weather on Disposition Effect
4.2The effect of weather on Disposition Effect
We examined how weather influences the disposition effect (DE) through its effect on risk preference, which is measured by PGR and PLR. We orderly regressed weather to PGR, weather to PLR, and weather to DE with different samples, including domestic corporations, domestic individuals, domestic institutions, and foreign institutions.
Because weather information is the monthly data (presented in the Table 2), we first separately sum every investor’s PGR, PLR, and DE of each transaction monthly according to four types of investors (presented in the Table 3).
We then ran the correlation coefficient analysis on the independent variables (weather), and found there is high correlation among independent variables. The results are presented in the table 5. From table 5, we can see that there is 59% positive correlation between temperature and rain, while there is 72% positive correlation between temperature and sunshine. Because there is high correlation among independent variables, we put each independent variable (temperature, rain, and sunshine) respectively in the regression analysis without putting all the three independent variables simultaneously in order to prevent collinearity problem. The regression results are presented in the Table 6, Table 7, Table 8, and Table 9.
Table 5 Correlation coefficients of independent variables
Temperature Rain Sunshine
Temperature 1 0.592 0.723
‧
In Table 6, we present the effect of weather on the disposition effect (DE) through its effect on PGR and PLR among domestic corporations.
The positive relationship between temperature and disposition effect is significant at 5%. The temperature is positive relative to PGR, PLR, and DE. The results are consistent with our expectations. When the temperature increases, domestic corporations exhibit risk aversion. Believe information has reflected in prices and tend to realize gains and losses, leading to both PGR and PLR increase. Because of the asymmetric risk aversion degree in gains and losses, the increasing degree of PGR is more than the increasing degree of PLR, leading to DE increase. On the other hand, when the temperature decreases, domestic corporations exhibit risk seeking. Believe information has not reflected in prices and rend to hold gains and losses, leading to both PGR and PLR decrease. Because of asymmetric risk seeking degree in gains and losses, the decreasing degree of PGR is more than that of PLR, which leads to DE decrease.
The positive relationship between the rain and disposition effect is not significant.
The rain is positive relative to PGR, PLR, and DE. The results are consistent with our expectations. When the rain increases, both PGR and PLR will increase. The increasing degree of PGR is more than increasing degree of PLR, leading to DE increase. On the other hand, when the rain decreases, both PGR and PLR will decrease. The decreasing degree of PGR is more than the decreasing degree of PLR, which leads to DE decrease.
The positive relationship between the sunshine and disposition effect is not significant. The sunshine is positive relative to PGR, negative relative to PLR, and positive relative to DE. The results are not consistent with our expectations. When the sunshine increases, PGR will increase but PLR will decrease, which leads to DE increase more. On the other hand, when the sunshine decreases, PGR will decrease, but PLR will increase, which leads to DE decrease more.
22
‧
國
立 政 治 大 學
‧
Na tiona
l Ch engchi University
Table 6 Regression analysis of domestic corporations
*indicates significant at 10%, **indicates significant at 5%, ***indicates significant at 1%. The regression analyses take the following forms.
PGR = α + β Temperature + ε PLR = α + β Temperature + ε DE = α + β Temperature + ε PGR = α + β Rain + ε PLR = α + β Rain + ε DE = α + β Rain + ε
PGR = α + β Sunshine + ε PLR = α + β Sunshine + ε DE = α + β Sunshine + ε
Model 1 2 3 Model 1 2 3 Model 1 2 3
Panel A:PGR Panel B:PLR Panel C:DE
Intercept 114.68 355.576*** 361.784*** Intercept 212.178 334.394*** 360.823*** Intercept -97.497 21.182 0.961
(0.77) (7.19) (3.66) (1.74) (8.91) (4.86) (-1.69) (1.04) (0.02)
Temperature 11.18 Temperature 5.692 Temperature 5.487**
(1.79) (1.12) (2.28)
Rain 0.089 Rain 0.047 Rain 0.042
(0.58) (0.4) (0.66)
Sunshine 0.138 Sunshine -0.123 Sunshine 0.262
(0.17) (-0.21) (0.82)
0.243 0.033 0.003 0.111 0.016 0.004 0.342 0.042 0.063
0.167 -0.064 -0.097 0.022 -0.082 -0.095 0.277 -0.054 -0.031
N 12 12 12 N 12 12 12 N 12 12 12
23 𝑅2
𝐴𝑑𝑗 − 𝑅2
𝑅2 𝑅2
𝐴𝑑𝑗 − 𝑅2 𝐴𝑑𝑗 − 𝑅2
‧
In Table 7, we present the effect of weather on the disposition effect (DE) through its effect on PGR and PLR among domestic individuals.
The positive relationship between the temperature and disposition effect is significant at 5%. The temperature is positive relative to PGR, PLR, and DE. The results are consistent with our expectations. When the temperature increases, domestic individuals exhibit risk aversion. Believe information has reflected in prices and tend to realize gains and losses, leading to both PGR and PLR increase. Because of the asymmetric risk aversion degree in gains and losses, the increasing degree of PGR is more than the increasing degree of PLR, leading to DE increase. On the other hand, when the temperature decreases, domestic individuals exhibit risk seeking. Believe information has not reflected in prices and tend to hold gains and losses, leading to both PGR and PLR decrease. Because of asymmetric risk seeking degree in gains and losses, the decreasing degree of PGR is more than that of PLR, which leads to DE decrease.
The positive relationship between rain and disposition effect is not significant. The rain is negative relative to PGR, negative relative to PLR, and positive relative to DE.
The results are not consistent with our expectations. When rain increases, both PGR and PLR will decrease. The decreasing degree of PGR is less than that of PLR, which leads to DE increase. On the other hand, when rain decreases, both PGR and PLR will increase. The increasing degree of PGR is less than that of PLR, leading to DE decrease.
The positive relationship between the sunshine and disposition effect is significant at 10%. The sunshine is positive relative to PGR, negative relative to PLR, and positive relative to DE. The results are not consistent with our expectations. When the sunshine increases, domestic individuals exhibit different risk attitude in gains and losses.
Domestic individuals exhibit risk aversion in gains, which leads to PGR increase, while exhibiting risk seeking in losses, leading to PLR decrease. Therefore, DE increase more.
24
‧
Table 7 Regression analysis of domestic individuals
*indicates significant at 10%, **indicates significant at 5%, ***indicates significant at 1%. The regression analyses take the following forms.
PGR = α + β Temperature + ε PLR = α + β Temperature + ε DE = α + β Temperature + ε
Intercept 66871 97687*** 91124*** Intercept 62129 79831*** 80054*** Intercept 4808.434 17870*** 11091**
(1.65) (8.23) (3.83) (1.65) (7.43) (3.7) (0.78) (8.1) (3.08)
Temperature 1071.31 Temperature 486.63 Temperature 582.207**
(0.63) (0.31) (2.26)
Rain -22.191 Rain -24.822 Rain 2.605
(-0.6) (-0.74) (0.38)
Sunshine 8.366 Sunshine -54.875 Sunshine 63.129*
(0.04) (-0.32) (2.18)
‧
In Table 8, we present the effect of weather on the disposition effect (DE) through its effect on PGR and PLR among domestic institutions.
The negative relationship between temperature and disposition effect is not significant. The temperature is negative relative to PGR, PLR, and DE. The results are not consistent with our expectations. When temperature increases, both PGR and PLR decrease. The decreasing degree of PGR is more than that of PLR, leading to DE decrease. On the other hand, when temperature decreases, both PGR and PLR increase.
The increasing degree of PGR is more than that of PLR, leading to DE increase.
The negative relationship between the rain and disposition effect is not significant.
The rain is positive relative to PGR, positive relative to PLR, and negative relative to DE. The results are consistent with our expectations. When the rain increases, both PGR and PLR will increase. The increasing degree of PGR is less than the increasing degree of PLR, which leads to DE decrease. On the other hand, when the rain decreases, both PGR and PLR will decrease. The decreasing degree of PGR is less than the decreasing degree of PLR, which leads to DE increase.
The negative relationship between the sunshine and disposition effect is not significant. The sunshine is negative relative to PGR, PLR, and DE. The results are consistent with our expectations. When the sunshine increases, both PGR and PLR will decrease. The decreasing degree of PGR is more than the decreasing degree of PLR, which leads to DE decrease. On the other hand, when the sunshine decreases, both PGR and PLR will increase. The increasing degree of PGR is more than the increasing degree of PLR, which leads to DE increase.
In summary, there is no significant relationship between weather and disposition effect of domestic institutions. A possible reason is that domestic institutions exhibit insignificant disposition effect, therefore weather may not affect their disposition effect.
26
‧
國
立 政 治 大 學
‧
Na tiona
l Ch engchi University
Table 8 Regression analysis of domestic institutions
*indicates significant at 10%, **indicates significant at 5%, ***indicates significant at 1%. The regression analyses take the following forms.
PGR = α + β Temperature + ε PLR = α + β Temperature + ε DE = α + β Temperature + ε PGR = α + β Rain + ε PLR = α + β Rain + ε DE = α + β Rain + ε
PGR = α + β Sunshine + ε PLR = α + β Sunshine + ε DE = α + β Sunshine + ε
27
Model 1 2 3 Model 1 2 3 Model 1 2 3
Panel A:PGR Panel B:PLR Panel C:DE
Intercept 191.755 102.751** 127.067 Intercept 172.133 98.424** 121.866 Intercept 19.623* 4.328 5.202
(1.43) (2.57) (1.62) (1.28) (2.47) (1.56) (1.93) (1.31) (0.79)
Temperature -3.762 Temperature -3.042 Temperature -0.72
(-0.67) (-0.54) (-1.69)
Rain 0.002 Rain 0.008 Rain -0.007
(0.01) (0.07) (-0.64)
Sunshine -0.203 Sunshine -0.182 Sunshine -0.022
(-0.32) (-0.29) (-0.41)
0.044 0.000 0.01 0.029 0.000 0.008 0.223 0.04 0.016
-0.052 -0.1 -0.089 -0.069 -0.01 -0.091 0.145 -0.056 -0.082
N 12 12 12 N 12 12 12 N 12 12 12
𝑅2 𝑅2 𝑅2
𝑅2 𝑅2 𝑅2
𝐴𝑑𝑗 − 𝑅2 𝐴𝑑𝑗 − 𝑅2 𝐴𝑑𝑗 − 𝑅2
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
In Table 9, we present the effect of weather on the disposition effect (DE) through its effect on PGR and PLR among foreign institutions.
The positive relationship between the temperature and disposition effect is not significant. The temperature is positive relative to PGR, PLR, and DE. The results are consistent with our expectations. When the temperature increases, both PGR and PLR will increase. The increasing degree of PGR is more than the increasing degree of PLR, which leads to DE increase. On the other hand, when the temperature decreases, both PGR and PLR will decrease. The decreasing degree of PGR is more than the decreasing degree of PLR, which leads to DE decrease.
The positive relationship between the rain and disposition effect is significant at 5%. The rain is positive relative to PGR, negative relative to PLR, and positive relative to DE. The results are not consistent with our expectations. When the rain increases, foreign institutions exhibit different risk attitude in gains and losses. Foreign institutions exhibit risk aversion in gains, which leads to PGR increase, while exhibiting risk seeking in losses, leading to PLR decrease. Therefore, DE increase more.
The positive relationship between the sunshine and disposition effect is not significant. The sunshine is positive relative to PGR, PLR, and DE. The results are not consistent with our expectations. When the sunshine increases, both PGR and PLR will increase. The increasing degree of PGR is more than the increasing degree of PLR, which leads to DE increase. On the other hand, when the sunshine decreases, both PGR and PLR will decrease. The decreasing degree of PGR is more than the decreasing degree of PLR, which leads to DE decrease.
28
‧
國
立 政 治 大 學
‧
Na tiona
l Ch engchi University
Table 9 Regression analysis of foreign institutions
*indicates significant at 10%, **indicates significant at 5%, ***indicates significant at 1%. The regression analyses take the following forms.
PGR = α + β Temperature + ε PLR = α + β Temperature + ε DE = α + β Temperature + ε PGR = α + β Rain + ε PLR = α + β Rain + ε DE = α + β Rain + ε
PGR = α + β Sunshine + ε PLR = α + β Sunshine + ε DE = α + β Sunshine + ε
29
Model 1 2 3 Model 1 2 3 Model 1 2 3
Panel A:PGR Panel B:PLR Panel C:DE
Intercept 169.264 454.973*** 397.164** Intercept 411.107* 497.303*** 363.085*** Intercept -241.843 -42.329 34.08
(0.73) (6.41) (2.79) (2.07) (9.68) (3.23) (-0.89) (-0.63) (0.21)
Temperature 14.733 Temperature 0.784 Temperature 13.949
(1.53) (0.09) (1.24)
Rain 0.244 Rain -0.269 Rain 0.514**
(1.1) (-1.68) (2.46)
Sunshine 1.012 Sunshine 0.565 Sunshine 0.448
(0.88) (0.62) (0.33)
0.189 0.109 0.072 0.001 0.221 0.038 0.133 0.376 0.011
0.108 0.02 -0.021 -0.1 0.143 -0.059 0.046 0.314 -0.088
N 12 12 12 N 12 12 12 N 12 12 12
𝑅2 𝑅2 𝑅2
𝑅2 𝑅2 𝑅2
𝐴𝑑𝑗 − 𝑅2 𝐴𝑑𝑗 − 𝑅2 𝐴𝑑𝑗 − 𝑅2
‧
methodology developed by Odean (1998) to measure disposition effect. We found that domestic individuals exhibit significant disposition effect and tend to realize winners too early and hold losers too long. This result is consistent with the findings from Choe and Eom (2009) and Li et al. (2013). However, the differences between our research and previous ones is that not only domestic individuals exhibit significant disposition effect, but also domestic corporations exhibit significant disposition effect. Besides, domestic individuals have higher disposition effect than domestic corporations. A possible explanation is that domestic individuals invest in the TAIFEX for the purpose of accumulating wealth. They avoid realizing losses because they feel regretful to see their wealth decreasing, while seeking gains realization because they feel pride to see their wealth increasing.We further examined how weather influences disposition effect of four types of investors, through its effect on risk attitude. We found that temperature is positive relative to PGR, PLR, and disposition effect of both domestic corporations and domestic individuals. The results are consistent with our expectations. When temperature increases, both domestic corporations and domestic individuals exhibit risk aversion. Believe information has reflected in prices and prices are too high now.
Investors tend to realize gains and losses, leading to both PGR and PLR increase.
Because of the asymmetric risk aversion degree in gains and losses, the increasing degree of PGR is more than the increasing degree of PLR, leading to disposition effect increase. On the other hand, when the temperature decreases, domestic corporations
30
‧
and domestic individuals exhibit risk seeking. Believe information has not reflected in prices and prices will continue to go up. Investors tend to hold gains and losses, which lead to both PGR and PLR decrease. Because of asymmetric risk seeking degree in gains and losses, the decreasing degree of PGR is more than the decreasing degree of PLR, which leads to disposition effect decrease.
Besides, the rain is positive relative to PGR, negative relative to PLR, and positive relative to disposition effect of foreign institutions. The results are not completely consistent with our expectations. We found that when rain increases, foreign institutions exhibit different risk attitude in gains and losses. Foreign institutions exhibit risk aversion in gains, which leads to PGR increase, while exhibiting risk seeking in losses, leading to PLR decrease. Therefore, DE increase more.
Moreover, the sunshine is positive relative to PGR, negative relative to PLR, and positive relative to disposition effect of domestic individuals. The results are not completely consistent with our expectations. When the sunshine increases, domestic individuals exhibit different risk attitude in gains and losses. Domestic individuals exhibit risk aversion in gains, which leads to PGR increase, while exhibiting risk seeking in losses, leading to PLR decrease. Therefore, DE increase more.
The empirical results indicated that weather influences disposition effect of different types of investors. However, there is no significant relationship between weather and disposition effect of domestic institutions. A possible explanation is that domestic institutions exhibit insignificant disposition effect, therefore weather may not influence their disposition effect.
In conclusion, we found that weather may influence disposition effect of different types of investors, through its effect on risk attitude. We recommended that further studies can lengthen the research period to investigate the research thoroughly.
31
‧
disposition effect: Large sample evidence across investor classes. International Review of Finance, 6(1-2), 43-78.(3) Cao, M. and Wei, J., (2004). Stock market returns: A note on temperature anomaly.
Journal of Banking and Finance, 29, 1559–1573.
(4) Chen, G., Kim, K. A., Nofsinger, J. R., and Rui, O. M., (2007). Trading performance, disposition effect, overconfidence, representativeness bias, and experience of emerging market investors. Journal of Behavioral Decision Making, 20(4), 425–
451.
(5) Choe, H. and Eom, Y., (2009). The disposition effect and investment performance in the futures market. Journal of Futures Markets, 29(6), 496-522.
(6) Fama, E. F., (1970). Efficient capital markets: A review of theory and empirical work. Journal of Finance, 31(1), 383-417.
(7) Frazzini, A., (2006). The disposition effect and underreaction to news. The Journal of Finance, 61(4), 2017-2046.
(8) Frino, A., Johnstone, D., and Zheng, H., (2004). The propensity for local traders in futures markets to ride losses: Evidence of irrational or rational behavior? Journal of Banking and Finance, 28(2), 353-372.
(9) Grinblatt, M. and Hany, B., (2001). The disposition effect and momentum, eScholarship University of California: Anderson Graduate School of Management-
32
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Finance UCLA.
(10) Hirshleifer, D. and Shumway, T., (2003). Good day sunshine: Stock returns and the weather. Journal of Finance, 58(3), 1009–1032.
(11) Kahneman. D. and Tversky, A., (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263-292.
(12) Li, H.C., Lin, C.H., Cheng, T.Y., and Lai, S., (2013). How different types of traders behavior in the Taiwan futures market. Journal of Futures Markets, 33(12), 1097-1117.
(13) Locke, P. R. and Mann, S. C., (2000). Do professional traders exhibit loss realization aversion? Working Paper.
(14) Mellers, B. A. and McGraw, A. P., (2001). Anticipated emotions as guides to choice. Current Directions in Psychological Science, 10(6), 210-214.
(15) Neumann, J. and Morgenstern, O., (1944). Theory of games and economic behavior, Princeton University: Princeton University.
(16) Odean, T., (1998). Are investors reluctant to realize their losses? Journal of Finance, 53(5), 1775-1798.
(17) Shefrin, H. and Statman, M., (1985). Disposition to sell winners too early and ride losers too long: Theory and evidence. Journal of Finance, 40(3), 777-790.
33