玉米区域试验中误差方差的异质性及其对品种评价的影响.pdfVIP

作物学报 ACTA AGRONOMICA SINICA 2013, 39(3): 449454 /zwxb/ ISSN 0496-3490; CODEN TSHPA9 E-mail: xbzw@ DOI: 10.3724/SP.J.1006.2013.00449 玉米区域试验中误差方差的异质性及其对品种评价的影响 王春平1,* 胡希远2,* 沈琨仑2 1 , 471003; 2 , 712100 : , 2003—2006 16 , , , , ; ; -, ; : ; ; ; Heterogeneity of Error Variance and Its Effect on Variety Evaluation in Corn Regional Trials WANG Chun-Ping 1,*, HU Xi-Yuan2,* , and SHEN Kun-Lun2 1 College of Agronomy, Henan University of Science and Technology, Luoyang 471003, China; 2 College of Agronomy, Northwest AF University, Yangling 712100, China Abstract: The objective of this study was to improve the method for statistical analysis in corn regional trials by insecting error variance. Based on the 16 data sets of the corn regional trials in the Northeast and North China from 2003 to 2006, we assessed the error variation between the environments and the fit-goodness of models for homogeneous and heterogeneous errors, and compared the statistical tests for trial effects from the two models. The results showed that the error variance largely varied be- tween environments in all of the considered trials. The model for heterogeneous errors fitted the trial data better than the model for homogeneous errors. Whether the heterogeneity of error variance was considered in the models considerably impacted the test about variety-environment interaction effects and little did the test about variety effects. The model for heterogeneous errors had higher test efficiency than