Models for Analyzing Economical Crop Yields in Thailand

Main Article Content

Pitsanu Tongkhow
Pornpit Sirima
Krisada Lekdee
Rachadasak Supengcum

บทคัดย่อ

This research is to propose an appropriate and efficient model for analyzing economical crop yields, rice, sugarcane and corn, in Thailand. The data were collected from provinces in the central region of Thailand, 22 provinces for rice, 9 provinces for corn and 14 provinces for sugarcanes, from 2007 to 2019. A linear mixed model (LMM) including spatial relationship explained by a conditional autoregressive (CAR) model was adopted. The Bayesian method was used for parameter estimation.  The factors related to the production of rice, sugarcane and corn were also investigated. The results showed that the amount of rain, temperature, region and spatial effects affected the yields.  When comparing the LMM with spatial relationship and the one without spatial relationship, it was found that the LMM model spatial relationship was more suitable than the one without spatial relationship. For each yield, the mean square error (MSE) of  the LMM was greater than the MSE of the proposed model (rice yields: 6.07 vs 1.59, corn yields: 7226.11 vs 6287.44 and sugarcane yields: 76.76 vs 46.69).

Article Details

บท
บทความวิจัย

References

The agricultural futures exchange of Thailand, the Thailand Futures Exchange (TFEX)[Internet]. [updated 2019 Aug 19; cited 2019 Aug 9]. Available from: https://www.tfex.co.th/th/about/glance.html.

Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986; 73:13-22.

Limmun W, Ingsrisawang, L. Study of factors affecting road traffic accidents using generalized estimating equations and generalized linear models. Journal of KMUTNB. 2010; 20(2): 311-21.

Lekdee K, Ingsrisawang L. Risk factors for malaria in Thailand using generalized estimating equations (GEE) and genalized linear mixed model (GLMM). J Health Sci. 2010; 19: 364-73.

Saengseedam P, Kantanantha N. Spatio-temporal model for crop yield forecasting. J Appl Stat. 2017; 44(3):427-40.

Sammatat, S. and Lekdee K. Estimation and Detection of Rice Yield in Thailand Using Spatial and Longitudinal Data Analysis. Applied Mathematical Sciences, 2019; 13(18): 877 - 84.

Bates DM, Mächler M, Bolker BM, et al. Fitting Linear Mixed-Effects Models Using lme4.2014[Internet]. [updated 2020 May 19; cited 2020 Jul 30]. Available from:http://arxiv.org/abs/1406.5823.

R Core Team. R: A Language and Environment for Statistical Computing. 2014[Internet]. [updated 2020 May 19; cited 2020 Jul 30] Available from: http://www.R-project.org

Magezi DA. Linear mixed-effects models for within-participant psychology experiments: an introductory tutorial and free, graphical user interface (LMMgui). Switzerland: University of Fribourg: Department of Medicine, Faculty of Science; 2015.

Pedroza CA. Bayesian forecasting model: predicting U.S. male mortality. Biostatistics. 2006; 7(4):530-50.

West BT, Welch KB, Galecki AT. Linear mixed models: A practical guide to using statistical software. NY : Chapman & Hall/CRC; 2007.

Banerjee S, Carlin BP, Gelfand AE. Hierarchical Modeling and Analysis for Spatial Data. Chapman and Hall/CRC Press. FL. Janeway CA, Travers P, Walport M, Shlomchik M. Immunobiology. 5th ed. New York: Garland Publishing; 2004.

Clayton D, Kaldor J. Empirical Bayes estimates of age-standardized relative risks for use in disease mapping. Biometrics. 1987;43:671-81.

Cowles MK. Applied Bayesian statistics: with R and OpenBUGS, New York: springer; 2013.

Baio G, Berardi A, Heath A. Bayesian cost-effectiveness analysis with the R package BCEA. New York: springer; 2017.

Temperature data. [Internet]. the thai meteorological department [updated 2019 Aug 19; cited 2019 Aug 9]. Available from: https://www.tmd.go.th/en/.

Yield data. [Internet]. the Office of Agricultural Economics [updated 2019 Aug 19; cited 2019 Aug 9]. Available from: http://www.oae.go.th/view/1/%E0%B8%AB%E0%B8%99%E0%B9%89%E0%B8%B2%E0%B9%81%E0%B8%A3%E0%B8%81/TH-TH.