한국 주식시장의 Fama-French 모형연구 = The Study of Fama-French Model on Korean Stock Market

다국어 초록 (Multilingual Abstract)

This paper investigates the importance of FF-3, 5 factor models, as well as FF-6 factor model by adding Liquidity(LIQ) to FF-5 factor model. Previous studies have emphasized the role of liquidity in Korean market. Using the monthly data from year 1998 to 2016, we have also divided the whole period into three different sub-periods; one before the global credit crisis, another during the crisis, and the last one after the crisis. We apply not only FF (2015, 2017) 4 metrics and spanning tests, but also Fama-MacBeth cross-sectional tests. According to FF (2015, 2017) 4 metrics, none of the models considered pass GRS test at 5% significance level, in the case of the whole period. Nevertheless, FF 3 factor model seems to perform the best, based on FF 4 metrics. When we replace HML by LIQ in FF-3 factor model, the original FF-3 factor model performs better. In addition, FF-5 factor model almost always perform better than FF-6 factor model. These tell us that LIQ may not be a necessary factor in Korea. When we look at sub-periods, most results are similar. Interestingly, most models also pass the GRS test in the case of sub-periods. Based on FF (2015, 2017) spanning test, we find Mkt, SMB, HML are not spanned, while RMW, CMA, LIQ are spanned, when we consider the whole period. The results are similar when we look at sub-periods, except that before the crisis, Mkt, SMB, HML factors all play roles, while after the crisis, only SMB plays a role. Overall, both FF (2015, 2017) 4 metrics and spanning tests show that FF 3 factor model is the best. The result of Fama-MacBeth cross-sectional test, however, doesn’t support it. As in previous studies, SMB plays a major role, and both LIQ and CMA tend to be statistically significant, which implies that FF 6 factor model performs relatively better. This is the case, whether we look at the whole period, or we look at sub-periods. To sum up, FF (2015, 2017) 4 metrics and spanning tests support FF-3 factor model, while Fama-MacBeth tests support FF-6 factor model.

참고문헌 (Reference)

1 윤상용, "한국 주식시장에서 유동성 요인을 포함한 3요인 모형의 설명력에 관한 연구" 한국재무학회 22 (22): 1-44, 2009

2 목남희, "잉여현금흐름과 투자성향이 기업가치에 미치는 영향" 한국상업교육학회 30 (30): 131-153, 2016

3 김주한, "금융위기와 위기지수를 활용한 국가별 분석" 한국상업교육학회 29 (29): 265-286, 2015

4 Novy-Marx, R., "The other side of value: The gross profitability premium" 108 (108): 1-28, 2013

5 Guo, B., "The five-factor asset pricing model tests for the Chinese stock market" 43 : 84-106, 2017

6 장운욱, "The Five-Factor Asset Pricing Model: Applications to the Korean Stock Market" 아시아.유럽미래학회 13 (13): 155-180, 2016

7 Serge R., "Re-examination of Fama-French Models in the Korean Stock Market" 26 : 23-45, 2019

8 Fama, E. F., "International tests of a five-factor asset pricing model" 123 : 441-463, 2017

9 Arnott R., "Fundamental Indexation" 61 : 83-99, 2004

10 Kim, S. H., "Evaluating asset pricing models in the Korean Stock market" 20 : 198-227, 2012

11 Hahn, J., "Determinants of the cross-sectional stock returns in Korea: evaluating recent empirical evidence" 38 : 88-106, 2016

12 Ball, R., "Deflating profitability" 117 (117): 225-248, 2015

13 Barillas, F., "Comparing Asset Pricing Models" NBER 2015

14 Fama, E. F., "Common risk factors in the returns on stocks and bonds" 33 (33): 3-56, 1993

15 Titman, S., "Capital investments and stock returns" 39 (39): 677-700, 2004

16 Ball, R., "An empirical evaluation of accounting income numbers" 159-178, 1968

17 Ball, R., "Accruals, cash flows, and operating profitability in the cross section of stock returns" 121 (121): 28-45, 2016

18 Fama, E. F., "A five-factor asset pricing model" 116 (116): 1-22, 2015

19 Gibbons, M., "A Test of the efficiency of a given portfolio" 57 (57): 1121-1152, 1989