This packages provides a statistical method to impute the missing values in accelerometer data. The methodology includes both parametric and semi-parametric multiple imputations under the zero-inflated poisson lognormal model. It also provides multiple functions to preprocess the accelerometer data previous to the missing data imputation. These includes detecting wearing and nonwearing time, selecting valid days and subjects, and creating plots.
Details
Package:
accelmissing
Type:
Package
Version:
1.0
Date:
2016-03-25
License:
GPL (>=2)
Author(s)
Jung Ae Lee <jungaeleeb@gmail.com>
References
Lee JA, Gill J (2016). Missing value imputation for physical activity data measured by accelerometer. Statistical Methods and Medical Research.