응용통계연구 (The Korean Journal of Applied Statistics)

  • 제37권6호
  • /
  • Pages.817-834
  • /
  • 2024
  • /
  • 1225-066X(pISSN)
  • /
  • 2383-5818(eISSN)
한국통계학회 (The Korean Statistical Society)
권호 표지
DOI QR코드

DOI QR Code

코스피 에너지 기업들의 로그수익률에 대한 독립성 검정과 분포 추론 연구

Independence test and distribution inference for log returns of KOSPI energy companies

  • 이유진 (가천대학교 응용통계학과) ;
  • 박소연 (가천대학교 응용통계학과) ;
  • 황은주 (가천대학교 응용통계학과)
  • Yujin Lee (Department of Applied Statistics, Gachon University) ;
  • Soyeon Park (Department of Applied Statistics, Gachon University) ;
  • Eunju Hwang (Department of Applied Statistics, Gachon University)
  • 투고 : 2024.02.11
  • 심사 : 2024.05.27
  • 발행 : 2024.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

에너지 산업은 개인의 일상 생활 뿐만 아니라 모든 분야의 국가 발전에 없어서는 안될 중요한 요소이다. 본 논문은 KOPSI에 상장된 상위6개의 에너지 기업에 대해, 주가 로그수익률을 이용한 독립성 검정과 수익률 분포 추론 연구를 목표로 한다. 독립성 검정을 위해 에너지 기업들의 조합을 이용한 교차분석을 시행한다. 수익률 분포를 탐구하고자, 정규분포와 지수분포를 포괄적으로 포함하는 압축지수분포 함수를 채택한다. 경험적 확률밀도함수와 압축지수분포 함수의 평균제곱차를 최소화함으로써 수익률 분포의 모수를 결정한다. 분포 추정의 정교성을 확보하기 위하여 윌콕슨 부호 순위 검정을 통한 비대칭성 및 왜도를 확인하고, 대칭성을 민족하지 않는 기업에 대해서는 추가적으로 양과 음의 수익률 각각에 대한 비대칭 압축지수분포를 찾는다. 본 연구결과는 수익률에 대한 확률이론 기반의 정보와 함께 명확한 분석을 제공하는데 기여할 수 있다.

Energy industry is an essential factor not only in the lives of individuals but also in the national development of all fields. This paper aims to study the independence test and distribution of log returns for top 6 energy companies in KOSPI. A cross-analysis on combinations of the six energy companies is conducted for the independence test. The return distributions are explored by adopting compressed exponential distribution function, which is a role of bridge between the normal and exponential distributions. Optimal compressed parameters of the return distributions are determined by minimizing the mean square difference between the empirical density function and compressed exponential function. To access a refinement of the distribution, asymmetry or skewness are tested via the Wilcoxon signed rank test, and the asymmetric compressed exponential distributions are inferred on two sides of negative and nonnegative returns, respectively. The results of this work can help to provide an explicit analysis along with probabilistic information about the returns.

키워드

과제정보

본 연구는 한국연구재단(NRF-2023R1A2C1005395)의 지원을 받아 수행되었습니다.

참고문헌

  1. Borak S, Misiorek A, and Weron R (2011). Models for heavy-tailed asset returns. In P Cizek, WK Hardle, and R Weron (2nd eds), Statistical Tools for Finance and Insurance (pp. 21-56), Springer, NewYork.
  2. Duttilo P, Gattone SA, and Iannone B (2023). Mixtures of generalized normal distributions and EGARCH models to analysis returns and volatility of ESG and traditional investments, AStA Advances in Statisitcal Analysis, Available from: http://doi.org/10.1007/s10182-023-00487-7
  3. Dytso A, Bustin R, Poor HV, and Shamai S (2018). Analytical properties of generalized Gaussian distributions, Journal of Statistical Distributions and Applications, 5, 6.
  4. Fama EF (1965). The behavior of stock market prices, The Journal of Business, 38, 34-105.
  5. Hayter A (2007). Probability and Statistics for Engineers and Scientists (4th ed), Thomson Brooks/Cole, Belmont.
  6. Hong C and Kwon T (2010). Distribution fitting for the rate of return and value at risk, Journal of the Korean Data Information Science Society, 21, 219-229.
  7. Hwang E (2023). Does COVID-19 affect the growth of the fourth industrial revolution?, Global Economic Review, 52, 220-234.
  8. Kim T and Song S (2011). Value-at-Risk estimation using NIG and VG distributions, Journal of the Korean Data Analysis Society, 13, 1775-1788.
  9. Marsaglia G, Tsang WW, and Wang J (2003). "Evaluating Kolmogorov's distribution", Journal of Statistical Software, 8, 1-4.
  10. McCauley JL (2004). Dynamics of markets: Econophysics and Finance, Cambridge University Press, Cambridge.
  11. Min S (2015). Goodness of fit and independence tests for major 8 companies of Korean stock market, The Korean Journal of Applied Statistics, 28, 1245-1255.
  12. Niederhoffer V and Osborne MFM (1966). Market making and reversal on the stock exchange, Journal of the American Statistical Association, 61, 897-916.
  13. Ryu CS, Lee SH, and Lee KK (1999). Multiple target angle tracking algorithm using angular innovations extracted from signal subspace, Electronics Letters, 35, 1520-1522.
  14. Seo J (2017). The effects of supply and demand shocks of oil price on the stock returns of energy firms: Evidence from firms listed in KOSPI, Korean Energy Economic Review, 16, 191-214.
  15. Working H (1960). Note on the correlation of first differences of averages in a random chain, Econometrica, 28, 916-918.
  16. Yoon J and Song S (2016). A numerical study of adjusted parameter estimation in the normal inverse Gaussian distribution, The Korean Journal of Applied Statistics, 29, 741-752.
  17. Yu S and Hwang E (2021). Time series analysis for Korean COVID-19 confirmed cases: HAR-TP-T model approach, The Korean Journal of Applied Statistics, 34, 239-254.