The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Bayesian structural equation modeling for analysis of climate effect on whole crop barley yield

청보리 생산량의 기후요인 분석을 위한 베이지안 구조방정식 모형

  • Kim, Moonju (Department of Statistics, Kangwon National University) ;
  • Jeon, Minhee (Department of Statistics, Kangwon National University) ;
  • Sung, Kyung-Il (Department of Feed Science and Technology, Kangwon National University) ;
  • Kim, Young-Ju (Department of Statistics, Kangwon National University)
  • 김문주 (강원대학교 정보통계학과) ;
  • 전민희 (강원대학교 정보통계학과) ;
  • 성경일 (강원대학교 사료생산과학전공) ;
  • 김영주 (강원대학교 정보통계학과)
  • Received : 2015.11.12
  • Accepted : 2016.01.28
  • Published : 2016.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

Whole Crop Barley (WCB) is a representative self-sufficient winter annual forage crop, along with Italian Ryegrass (IRG), in Korea. In this study, we examined the path relationship between WCB yield and climate factors such as temperature, precipitation, and sunshine duration using a structural equation model. A Bayesian approach was considered to overcome the limitations of the small WCB sample size. As prior distribution of parameters in Bayesian method, standard normal distribution, the posterior result of structural equation model for WCB, and the posterior result of structural equation model for IRG (which is the most popular winter crop) were used. Also, Heywood case correction in prior distribution was considered to obtain the posterior distribution of parameters; in addition, the best prior to fit the characteristics of winter crops was identified. In our analysis, we found that the best prior was set by using the results of a structural equation model to IRG with Heywood case correction. This result can provide an alternative for research on forage crops that have hard to collect sample data.

청보리는 국내에서 자급자족되는 중요한 동계 풀사료이다. 본 연구는 구조방정식 모형을 이용하여 온도, 강수 및 일조시간과 관련 있는 기후요인이 청보리의 생산량에 미치는 경로와 영향력을 파악하였다. 청보리의 소표본 자료의 한계를 보완하기 위하여 베이지안 구조방정식 방법을 사용하였다. 베이지안 방법의 사전분포로 표준정규분포, 청보리 자료의 빈도론적 구조방정식 결과와 가장 대중적인 동계 풀사료인 이탈리안 라이그라스의 빈도론적 구조방정식 결과를 이용하였다. 또한, 사전분포의 헤이우드 케이스 수정을 하지 않은 경우와 수정한 경우에 대하여 구한 사후분포의 결과를 비교하여 동계작물의 생육특성과 잘 부합하는 사전분포를 탐색하였다. 분석 결과, 사전분포의 헤이우드 케이스를 수정하여 이탈리안 라이그라스의 빈도론적 구조방정식 결과를 사전분포로 사용하는 것이 가장 적절하였다. 그러므로 본 연구의 베이지안 접근법을 통해 표본 수집이 어려운 풀사료 연구에 좋은 제안이 될 것이다.

Keywords

References

  1. Arbuckle, J. (2005). Amos 6.0 user's Guide, Marketing Department, SPSS incorporated, Chicago.
  2. Bollen, K. A. (1987). Total, direct, and indirect effects in structural equation models, Sociological methodology, 17, 37-69. https://doi.org/10.2307/271028
  3. Boomsma, A. (1983). On the robustness of LISREL (maximum likelihood estimation) against small sample size and non-normality (Doctoral dissertation, University of Groningen).
  4. Descheemaeker, K., Smith, A. P., Robertson, M. J., Whitbread, A. M., Huth, N. I., Davoren, W., and Llewellyn, R. (2014). Simulation of water-limited growth of the forage shrub saltbush(Atriplex nummularia Lindl.) in a low-rainfall environment of southern Australia, Crop and Pasture Science, 65, 1068-1083. https://doi.org/10.1071/CP13452
  5. Dunson, D. B., Palomo, J., and Bollen, K. (2005). Bayesian structural equation modeling, (Technical report).
  6. Gao, S., Mokhtarian, P. L., and Johnston, R. A. (2008). Nonnormality of data in structural equation models, Transportation Research Record, Journal of the Transportation Research Board, 2082, 116-124. https://doi.org/10.3141/2082-14
  7. Kazama, J. I., De Saeger, D., Kuroda, K., Murata, M., and Torisawa, K. (2010). A Bayesian method for robust estimation of distributional similarities, In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, 247-256.
  8. Kim, J. D., Lee, H. J., Jeon, K. H., Yang, G. Y, Kwon, C. H., Sung, H. G., Hwangbo, S., and Jo, I. H. (2010). Effect of harvest stage, Wilting and Crushed rice on the forage production and silage quality of organic Whole crop barely, Journal of The Korean Society of Grassland and Forage Science, 30, 25-34. https://doi.org/10.5333/KGFS.2010.30.1.025
  9. Kim, K. D., Sung, K. I., Jung, Y. S., Lee, H. I., Kim, E. J., Nejad, J. G., Jo, M. H., and Lim, Y. C. (2013). Suitability classes for Italian ryegrass(Lolium multiflorum Lam.) using soil and climate digital database in Gangwon provice, Journal of the Korean Society of Grassland and Forage Science, 32, 437-446.
  10. Kim, M. J., Sung, K. I., Kim, B. W., Peng, J. L., Hyeon, D. H., Lee, B. H., Kim, E. J., Jo, M. H., Lim, Y. C., and Kim, K. D. (2014a). Prediction of Italian ryegrass yield affecting climate and soil, In Proceedings of 2014 Annual Congress of Korean Society of Animal Sciences and Technology, 204.
  11. Kim, M. J., Sung, K. I., and Kim, Y. J. (2014b). Analysis of climate effects on Italian ryegrass yield via structural equation model, The Korean Journal of Applied Statistics, 27, 111-120.
  12. Kim, M. J., Sung, K. I., and Kim, Y. J. (2015). Analysis of causal relation for climates and yield of Italian ryegrass and Whole crop barley via structural equation model(Manuscript).
  13. Kolenikov, S. and Bollen, K. A. (2012). Testing negative error variances is a Heywood case a symptom of misspecification, Sociological Method and Research, 41, 124-167. https://doi.org/10.1177/0049124112442138
  14. Lee, B. H., Kim, Sung, K. I., Kim, E. J., Peng, J. L., Ji, D. H., Ghassemi Nejad, J., Kim, M. J., Jo, M. H., Lim, Y. C., and Kim, K. D. (2013). Comparison between groups variables influencing dry matter yield of Whole crop corn, In Proceeding of The Korean Society of Grassland and Forage Science.
  15. Lim, S. and Melville, N. P. (2009). Robustness of structural equation modeling to distributional misspecification: empirical evidence and research guidelines, Electronic copy available at http://ssm.com/abstract=1375251.
  16. Mullen, M. R., Milne, G. R., and Doney, P. M. (1995). An international marketing application of outlier analysis for structural equations: a methodological note, Journal of International Marketing, 3, 45-62.
  17. Peng, J. L., Kim, E. J., Lee, B. H., Ji, D. H., Ghassemi Nejad, J., Kim, M. J., Jo, M. H., Lim, Y. C., and Kim, K. D. (2013). Comparison of annual dry matter yields of whole crop corn, In Proceeding of The Korean Society of Grassland and Forage Science.
  18. Rindskopf, D. (1984). Using phantom and imaginary latent variables to parameterize constraints in linear structural models, Psychometricka, 49, 37-47. https://doi.org/10.1007/BF02294204
  19. Schlenker, W. and Roberts, M. J. (2009). Nonlinear temperature effects indicate severe damages to U.S. crop yield under climate change, In Proceeding of the National Academy of Sciences, 106, 15594-15598. https://doi.org/10.1073/pnas.0906865106
  20. Shin, C. N., Ko, K. H., and Kim, J. D. (2012). Agronomic characteristics and forage productivity of Italian ryegrass(Lolium multiorum Lam.) cultivar, Journal of The Korean Society of Grassland and Forage Science, 32, 229-236. https://doi.org/10.5333/KGFS.2012.32.3.229
  21. Sung, K. I., Kim, B. W., Peng, J. L., Hyeon, D. H., Lee, B. H., Kim, E. J., Jo, M. H., Lim, Y. C., Kim, K. D., and Kim, M. J. (2014). Prediction of Italian ryegrass yield affecting climate variables, In Proceedings of 2014 Annual Congress of Korean Society of Animal Sciences and Technology, 203.
  22. Wothke, W. (1993). Nonpositive definite matrices in structural modeling, In Testing Structure Equation Models, edited by K.A.Bollen and J.S. Ling. Newbury Park: Sage, 256-256.

Cited by

  1. Yield modeling for prediction of regional whole-crop barley productivity pp.17446961, 2019, https://doi.org/10.1111/grs.12233