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

- Journal title : Korean Journal of Applied Statistics
- Volume 29, Issue 2, 2016, pp.331-344
- Publisher : The Korean Statistical Society
- DOI : 10.5351/KJAS.2016.29.2.331

Title & Authors

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

Kim, Moonju; Jeon, Minhee; Sung, Kyung-Il; Kim, Young-Ju;

Kim, Moonju; Jeon, Minhee; Sung, Kyung-Il; Kim, Young-Ju;

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.

Keywords

whole crop barley;structural equation model;Bayesian approach;Heywood cases;longitudinal data;

Language

Korean

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.

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.

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.

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.

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.

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.

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.

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.

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.