择时交易的小概率困境

Small probability dilemma of market-timing

本文通过研究我国从上世纪90年代初至2010年1月的上证综指和深证成指,发现极少数具有超常收益的交易日对股票市场的长期收益具有显著的影响:如果错过上证综指涨幅最大的10个交易日,以上证综指作为被动投资组合的收益将会降低92%之多;深证市场也有类似结果。由于10个交易日只占整个样本时期的0.2%,通过择时找准这10天从概率上来说是不可能的,理性的投资者应该放弃择时而选择长期投资,同时这个结论也间接地支持有效市场假说。

From the early 1990’s to January 2010, the long-term returns from both the Shanghai Composite Index （ShCI） and the Shenzhen Component Index are found to be affected tremendously by a few trading days with extreme returns. For example, the return on ShCI would be reduced by 92%, if one took out the ten trading days with the highest returns. Similar results are in the Shenzhen market as well. Comparing with the whole period covered in this paper, ten days amount to a tiny 0.2%, so any attempt to correctly time those ten days appears to be extremely difficult, if not downright impossible. Therefore, rational investors should choose long-term investing instead of market-timing. The result of this paper also seems to support EMH indirectly.