최적 포트폴리오의 개선: 새로운 상관행렬 개선방법 = Improvement of Optimal Portfolio using Correlation Matrix

다국어 초록 (Multilingual Abstract)

This study empirically investigates the practical applicability of Markowitz (1952) optimization function through correlation matrixes among stocks in the Korean stock market. The proposed method is a correlation matrix that removes the property of a market factor included in the sample correlation matrix, that is, the non-market correlation matrix. For comparison with the previous studies, the correlation matrixes that is known in the optimal portfolio are utilized, along with equal-weighted and value-weighted portfolios. According to the results, the optimal portfolio from the non-market correlation matrix may construct better diversified portfolio, and then, achieve lower risk and higher performance compared to other correlation matrixes. In comparison by the perspective of prediction error from expected return and standard deviation of returns, moreover, the optimal portfolio from the non-market correlation matrix has much lower magnitude of sensitivity from prediction error of input variables than those from the other correlation matrixes. These results show the evidence supporting that the non-market correlation matrix has a comparative benefit for improving the practical applicability of the optimal function as well as for effectively reducing the influence of prediction error from input variables.

국문 초록 (Abstract)

본 연구는 상관행렬 추정방법을 이용한 최적 포트폴리오의 개선 여부를 조사하였다. 제안하는 상관행렬은 표본상관행렬에 포함된 시장요인의 속성을 제거한 상관행렬(비시장상관행렬)이다. 비교대상은 표본상관행렬과 함께 기존연구에서 개선효과를 갖는 상관행렬들(일정상관행렬, 시장요인 상관행렬, 3요인 상관행렬, shrinkage 상관행렬)로부터의 최적 포트폴리오, 그리고 동일가중 및 가치가중 포트폴리오이다. 검증결과에 의하면, 표본상관행렬을 이용한 최적 포트폴리오는 이론적으로 기대하는 잘 분산 투자된 포트폴리오를 구성하지 못한다. 비시장상관행렬을 이용한 최적 포트폴리오는 다른 상관행렬과 비교하여 보다 잘 분산 투자된 포트폴리오를 구성하고, 이를 통해 낮은 위험과 높은 투자성과를 달성한다. 더욱이 입력변수(기대수익, 표준편차) 예측오류에 대하여, 비시장상관행렬로부터의 최적 포트폴리오는 다른 상관행렬에 비교하여 현저히 작은 민감도를 보인다. 또한, 실증분석의 변경에 따른 강건성 검증에서도 동일한 증거를 보였다.

참고문헌 (Reference)

1 김인수, "우리나라 주식시장에서의 주식프리미엄 퍼즐에 관한 연구" 한국재무학회 21 (21): 1-32, 2008

2 엄철준, "시장상황을 고려한 기대 주식수익률의 횡단면에 관한 재조사" 한국재무학회 25 (25): 599-639, 2012

3 Eom, C., "Two-faced Property of a Market Factor in Asset Pricing and Diversification Effect" 471 : 190-199, 2017

4 Brown, S., "The Number of Factors in Security Returns" 44 (44): 1247-1262, 1989

5 Michaud, R., "The Markowitz Optimization Enigma : Is ’Optimized’Optimal?" 45 (45): 31-42, 1989

6 Markowitz H., "The Early History of Portfolio Theory : 1600-1960" 55 (55): 5-16, 1999

7 Fabozzi, F., "Robust Portfolio Optimization and Management" John Wiley & Sons 2007

8 Plerou, "Random matrix approach to cross correlations in financial data" 65 : 066126-, 2002

9 Markowitz H., "Portfolio Selection" 7 (7): 77-91, 1952

10 DeMiguel V., "Optimal versus Naive Diversification : How Inefficient is the 1/N Portfolio Strategy?" 22 : 1915-1953, 2009

11 Best, M., "On the Sensitivity of the Mean-Variance Efficient Portfolio to Changes in Asset Means : Son Analytical and Computational Results" 4 (4): 315-342, 1991

12 Trzcinka, C., "On the Number of Factors in the Arbitrage Pricing Model" 41 (41): 347-368, 1986

13 Chan, L., "On Portfolio Optimization : Forecasting Covariances and Choosing the Risk Model" 12 (12): 937-974, 1999

14 Rubinstein, M., "Markowitz’s “Portfolio Selection”: A Fifty-Year Retrospective" 57 (57): 1041-1045, 2002

15 Duchin, R., "Markowitz Versus the Talmudic Portfolio Diversification Strategies" 71-74, 2009

16 Tu, J., "Markowitz Meets Talmud : A Combination of Sophisticated and Naive Diversification Strategies" 99 : 204-2015, 2011

17 King, B., "Market and Industry Factors in Stock Price Behavior" 39 (39): 139-190, 1966

18 Ledoit, O., "Improved estimation of the covariance matrix of stock returns with an application to portfolio selection" 10 : 603-621, 2003

19 Elton, E. M., "Improved Estimates of Correlation Coefficients and Their Impact on Optimum Portfolios" 12 (12): 303-318, 2006

20 Ledoit, O., "Honey, I shrunk the sample covariance matrix" 31 : 110-119, 2004

21 Kleibergen, F., "Generalized reduced rank tests using the singular value decomposition" 133 : 97-126, 2006

22 Markowitz H., "Foundations of Portfolio Theory" 46 (46): 469-477, 1991

23 Elton, E., "Estimating the Dependence Structure of Share Prices : Implications for Portfolio Selection" 28 : 1203-1232, 1973

24 Fama, E., "Efficient Capital Markets : II" 46 (46): 1575-1617, 1991

25 Fama, E., "Efficient Capital Markets : A Review of Theory and Empirical Work" 25 (25): 383-417, 1970

26 Eom, C., "Effects of the Market Factor on portfolio Diversification : the case of Market Crashes" 44 (44): 71-83, 2015

27 Eom, C., "Effects of Common Factors on Stock Correlation Networks and Portfolio Diversification" 49 : 1-11, 2017

28 Eom, C., "Effect of Changing Data Size on Eigenvalues in the Korean and Japanese Stock Markets" 388 : 4780-4786, 2009

29 Sengupta, A., "Distributions of singular values for some random matrices" 60 : 3389-, 1999

30 Brinson, G., "Determinants of portfolio Performance Ⅱ : An Update" 47 (47): 40-48, 1991

31 Brinson, G., "Determinants of portfolio Performance" 42 (42): 133-138, 1986

32 Fama, E., "Common Risk Factors in the Returns on Stocks and Bonds" 33 : 3-56, 1993

33 Elton, E., "Are betas best?" 23 : 1375-1384, 1979

34 Roll, R., "An Empirical Investigation of the Arbitrage Pricing Theory" 35 (35): 1073-1103, 1980

35 Merton, R., "An Analytic Derivation of the Efficient Portfolio Frontier" 7 (7): 1851-1872, 1972

36 Sharpe, W., "A simplified model for portfolio analysis" 9 : 277-293, 1963

37 Connor, G., "A Test for the Number of Factors in an Approximate Factor Model" 48 (48): 1263-1291, 1993

38 Eom, C., "A New Method for Better Portfolio Investment : A Case of the Korean Stock market" 49 : 213-231, 2018