توجه: محتویات این صفحه به صورت خودکار پردازش شده و مقاله‌های نویسندگانی با تشابه اسمی، همگی در بخش یکسان نمایش داده می‌شوند.
۱Handling Missing at Random Data: A Case Study of Imputation Methods
نویسنده(ها): ، ،
اطلاعات انتشار: World Applied Sciences Journal، بيست و ششم،شماره۸، ۲۰۱۳، سال
تعداد صفحات: ۷
Missing data is one of the most important problems which needs to be addressed in longitudinal studies. The method to handle this problem depends on the dropout mechanism. In most studies, the researchers accept this mechanism as random and apply imputation methods to deal with the missing data in statistical analysis. The main issue in this paper is to estimate the mean dosage of methadone at which patients continue their treatment to reduce drug use. In this study, the patients were treated at three methadone dose levels. However, a major percentage of them (36%) did not return to the treatment practice at the last time. Some important covariates are used in the multiple imputation models to obtain better estimates. Since the parameters of the distribution in the multiple imputation methods are appropriately estimated because of accounting for the uncertainty due to the imputation, an estimate with less bias and more realistic standard error is produced by multiple imputation methods. A few important imputation methods are applied to estimate the mean dose level of methadone in the last time, including missing data.
نمایش نتایج ۱ تا ۱ از میان ۱ نتیجه