New insights on ''A semi-parametric model for wearable sensor-based physical activity monitoring data with informative device wear"
碩士 === 國立中央大學 === 統計研究所 === 107 === In this article we indicate and solve two issues found in Song et al.(2018). First, we characterize the time-scale invariant property of the semi-parametric model they utilize, and show that their R package provides different results after time scaling. Second, we...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2019
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/5xfabt |
| Summary: | 碩士 === 國立中央大學 === 統計研究所 === 107 === In this article we indicate and solve two issues found in Song et al.(2018). First, we characterize the time-scale invariant property of the semi-parametric model they utilize, and show that their R package provides different results after time scaling. Second, we find the setting of their main simulation does not follow the assumptions of the utilized model, which may lead to an unreliable conclusion. We provide corrected R code for practical use, and give an appropriate simulation setting to illustrate the behavior of the utilized model in a real-data example.
|
|---|