Search Results
Author: Reiter, Jerome P.
Resulting in 2 citations.
| 1. |
Hill, Jennifer L. Reiter, Jerome P. Zanutto, Elaine L. |
|
A Comparison of Experimental and Observational Data Analyses In: Applied Bayesian Modeling and Causal Inference from Incomplete-Data Perspectives. A. Gelman and X. Meng, eds., New York: Wiley, 2007: 49-60 Cohort(s): Children of the NLSY79 Publisher: Wiley Online Keyword(s): Birthweight; Child Care; I.Q.; Missing Data/Imputation; Peabody Picture Vocabulary Test (PPVT); Propensity Scores; Test Scores/Test theory/IRT Permission to reprint the abstract has not been received from the publisher. In this paper, we illustrate the potential efficacy of these types of analyses. The causal question we address concerns the effects on intelligence test scores of a particular intervention that provided very high quality childcare for children with low birth weights.We have data from the randomized experiment performed to evaluate the causal effect of this intervention, as well as observational data from the National Longitudinal Survey of Youth on children not exposed to the intervention. Using these two datasets, we compare several estimates of the treatment effect from the observational data to the estimate of the treatment effect from the experiment, which we treat as the gold standard. ...We also demonstrate the use of propensity scores with data that has been multiply imputed to handle pretreatment and post-treatment missingness. To our knowledge, these other constructed observational studies performed analyses using only units with fully observed data. |
|
|
Bibliography Citation
Hill, Jennifer L., Jerome P. Reiter and Elaine L. Zanutto. "A Comparison of Experimental and Observational Data Analyses" In: Applied Bayesian Modeling and Causal Inference from Incomplete-Data Perspectives. A. Gelman and X. Meng, eds., New York: Wiley, 2007: 49-60
|
| 2. |
Hu, Jingchen Mitra, Robin Reiter, Jerome P. |
|
Are Independent Parameter Draws Necessary for Multiple Imputation? The American Statistician 67,3 (2013): 143-149. Also: http://www.tandfonline.com/doi/full/10.1080/00031305.2013.821953#.UjISYneHp4N Cohort(s): Children of the NLSY79, NLSY79 Publisher: American Statistical Association Keyword(s): Breastfeeding; Missing Data/Imputation; Peabody Individual Achievement Test (PIAT- Math); Statistical Analysis In typical implementations of multiple imputation for missing data, analysts create m completed datasets based on approximately independent draws of imputation model parameters. We use theoretical arguments and simulations to show that, provided m is large, the use of independent draws is not necessary. In fact, appropriate use of dependent draws can improve precision relative to the use of independent draws. It also eliminates the sometimes difficult task of obtaining independent draws; for example, in fully Bayesian imputation models based on MCMC, analysts can avoid the search for a subsampling interval that ensures approximately independent draws for all parameters. We illustrate the use of dependent draws in multiple imputation with a study of the effect of breast feeding on children’s later cognitive abilities. |
|
|
Bibliography Citation
Hu, Jingchen, Robin Mitra and Jerome P. Reiter. "Are Independent Parameter Draws Necessary for Multiple Imputation?" The American Statistician 67,3 (2013): 143-149.
|