空氣污染對投資者行為之影響

本研究從投資人角度，探討空氣污染是否為影響股市投資者情緒之潛在因子，研究期間為 2009 年 1 月 5 日至 2018 年 12 月 28 日。本研究採用Logit與二階最小平方估計法(2SLS)建構資產報酬率的波動模型。研究結果表明，在不同模型下之PM_2.5濃度均對日報酬並無顯著關係。Logit模型中，溫度在10%與5%的顯著水準下與日報酬呈現負相關；二階最小平方估計法(SO_2)中，溫度在5%的顯著水準下與日報酬呈現負相關，以上表明溫度越高將顯著拉低投資人績效表現，並且也與過往文獻結論一致。本研究結果可提供量化投資者與金融機構研究人員，在進行模型架構與風險管理策略研擬之參考。然而，即使目前無證據顯示PM_2.5將影響台灣資本市場運作，但國外且尤其污染嚴重之地區皆顯示空氣污染與股市報酬率間的負向關係，因此可藉由此實證結果提醒投資人與政府，及早預防與遏止空氣污染程度持續惡化，因空氣汙染不僅僅正在危害大家的生存環境，更有可能正在無形中侵蝕著的獲利成果。

關鍵字

空氣污染；行為財務學； Logit模型； 2SLS

並列摘要

This study takes the perspective of investors and examines whether air pollution is a potential factor affecting investor sentiment in the stock market, and the research period is from January 5, 2009 to December 28, 2018. This study uses the Logit model and two-stage least squares estimation method (2SLS) to construct fluctuation models of return on assets. The results show that PM_2.5 concentrations are not significantly related to daily returns in different models. In the Logit model, temperatures are negatively correlated to daily returns at significance levels of 10% and 5%. In the 2SLS method (SO_2), the temperature is negatively correlated to daily returns at a significance level of 5%. The above results show that higher temperatures will significantly lower investor performance are consistent with previous literature findings. The results of this study provide a reference for quantitative investors and financial institution researchers in the development of model architecture and risk management strategies. However, even if there is no evidence to support that PM_2.5 will affect the operation of Taiwan’s capital market, in foreign countries, especially areas with severe pollution, it is suggested that there is a negative relationship between air pollution and stock market returns. Therefore, this empirical result can remind investors and governments to prevent and curb the continued deterioration of air pollution as it is not only endangering the environment, it may also implicitly dampen profitability.