Weighted TVEM SAS Macro Users’ Guide (Version 2.6)

Citation

Dziak, John J.; Li, Runze; & Wagner, Aaron T. (2017). Weighted TVEM SAS Macro Users’ Guide (Version 2.6). The Methodology Center, Penn State.

Abstract

The WeightedTVEM macro was developed by the Methodology Center for fitting time-varying effects models (TVEMs; see Hastie & Tibshirani, 1993; Shiyko, Lanza, Tan, Li, & Shiffman, 2012; Tan, Shiyko, Li, Li, & Dierker, 2012) on complex datasets, such as Add Health, a nationally representative survey dataset (see Chen & Chantala, 2014). Such datasets may involve clustering (e.g., students are nested within schools) and survey weights (e.g., participants represent different numbers of population members due to systematically unequal probabilities of selection). Datasets may be cross-sectional (e.g., one observation per student) or longitudinal (e.g., multiple related observations per student over time). This brief users’ guide assumes familiarity with TVEMs and with the original TVEM macro without survey weights (TVEM SAS Macro, version 3.1, 2015; see Li, Dziak, Tan, Huang, Wagner, & Yang 2015). Below we explain how the %WeightedTVEM macro differs from the usual macro.

URL

https://methodology.psu.edu/sites/default/files/software/tvem/WeightedTVEMUsersGuidev2.6.pdf

Reference Type

Report

Author(s)

Dziak, John J.
Li, Runze
Wagner, Aaron T.

Year Published

2017

Publisher

The Methodology Center, Penn State

Reference ID

8157