주가 추세요인의 추출방법과 수익성 분석

This paper proposes a trend factor as a potential factor explaining the variation of stock returns. The trend factor is constructed by buying stocks with the highest expected returns and selling stocks with the lowest expected returns, where the expected returns for stocks are estimated using trend signals extracted from a total of 11 short-term and mid- to long-term moving average stock prices. The trend factor generates a significant excess return of 1.99% per month on average, which is even larger during the stock market recession, reaching an average of 2.21% per month. In addition, the trend factor has a dominant explanatory power on stock returns than other factors, such as short-term reversal factor, midterm momentum factor, and long-term reversal factor, which are well known as stock price trends in the existing literature, and is superior to the Fama-French three factors and momentum factor in terms of the Sharpe ratio and maximum drawdown. In the cross-sectional regression analysis, the trend factor maintains its explanatory power on stock returns even after controlling for the company size, market value to book value, past stock returns, and information uncertainty. The predictability of trend factors persists up to the next 6 months and remains significant even after changing the sample stocks, changing the method of calculating trend factors, and controlling various characteristic variables and information uncertainty variables that may affect the stock prices.