MULTIPLE IMPUTATION MIGAMMI ALGORITHM
Main Article Content
Abstract
Missing data are common in multi-environmental experiments however sophisticated they are. Thus, it is essential to use appropriate methods of analysis to reduce the impact generated by the loss of information. Data imputation consists in one of the most common techniques used to overcome the problem of missing values, it estimates missing data by plausible values; subsequently, the analyses are carried out on the complete data. This work aims to propose a new multiple imputation method for data from multi-environment trials, resulting from the proposal based on the simple residuals of a linear regression model. Alterations were made in the simple imputation algorithm EM-AMMI to accommodate the additive main effect and generalized multiplicative interaction GAMMI. The quality of the multiple imputations method was evaluated by using accurate general statistics distributions, which combines the variance among imputation and mean square deviation, and normalized root mean square error (NRMSE). For such, simulations of random values at levels of 10%, 20%, 30% and up to 40% were performed from two real data set and the obtained corresponding imputations. The overall mean accuracy and NRMSE results, given the low values obtained, considering the proposed method, demonstrate the high quality of the proposed multiple imputation algorithm MIGAMMI.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).