scholarly journals Modelling Error Dependence in Categorical Longitudinal Data

2021 ◽  
pp. 173-194
Author(s):  
Dimitris Pavlopoulos ◽  
Paulina Pankowska ◽  
Bart Bakker ◽  
Daniel Oberski
Keyword(s):  
Measurement Error ◽  
Longitudinal Data ◽  
Gold Standard ◽  
Markov Models ◽  
Hidden Markov ◽  
Register Data ◽  
Standard Data ◽  
Data Source ◽  
Error Dependence ◽  
Modelling Error

Hidden Markov models (HMMs) offer an attractive way of accounting and correcting for measurement error in longitudinal data as they do not require the use of a ‘gold standard’ data source as a benchmark. However, while the standard HMM assumes the errors to be independent or random, some common situations in survey and register data cause measurement error to be systematic. HMMs can correct for systematic error as well if the local independence assumption is relaxed. In this chapter, we present several (mixed) HMMs that relax this assumption with the use of two independent indicators for the variable of interest. Finally, we illustrate the results of some of these HMMs with the use of an example of employment mobility. For this purpose, we use linked survey-register data from the Netherlands.

scholarly journals Reconciliation of inconsistent data sources using hidden Markov models

2020 ◽  
Vol 36 (4) ◽  
pp. 1261-1279
Author(s):  
Paulina Pankowska ◽  
Dimitris Pavlopoulos ◽  
Bart Bakker ◽  
Daniel L. Oberski
Keyword(s):  
Measurement Error ◽  
Data Collection ◽  
Hidden Markov Models ◽  
Markov Models ◽  
Hidden Markov ◽  
Parameter Estimates ◽  
Collection Process ◽  
Data Collection Process ◽  
Linkage Error ◽  
Over Time

This paper discusses how National Statistical Institutes (NSI’s) can use hidden Markov models (HMMs) to produce consistent official statistics for categorical, longitudinal variables using inconsistent sources. Two main challenges are addressed: first, the reconciliation of inconsistent sources with multi-indicator HMMs requires linking the sources on the micro level. Such linkage might lead to bias due to linkage error. Second, applying and estimating HMMs regularly is a complicated and expensive procedure. Therefore, it is preferable to use the error parameter estimates as a correction factor for a number of years. However, this might lead to biased structural estimates if measurement error changes over time or if the data collection process changes. Our results on these issues are highly encouraging and imply that the suggested method is appropriate for NSI’s. Specifically, linkage error only leads to (substantial) bias in very extreme scenarios. Moreover, measurement error parameters are largely stable over time if no major changes in the data collection process occur. However, when a substantial change in the data collection process occurs, such as a switch from dependent (DI) to independent (INDI) interviewing, re-using measurement error estimates is not advisable.

Using Longitudinal Data to Model Teachers' Ratings of Classroom Behavior as a Dynamic Process

1981 ◽  
Vol 6 (3) ◽  
pp. 237-255 ◽  
Author(s):  
Ian Plewis
Keyword(s):  
Measurement Error ◽  
Longitudinal Data ◽  
Dynamic Process ◽  
Categorical Data ◽  
Continuous Time ◽  
Classroom Behavior ◽  
Markov Models ◽  
Time Measurement ◽  
Small Sample ◽  
Educational Implications

Simple Markov models are fitted to a small sample of longitudinal categorical data of teachers' ratings of children's classroom behavior. Although the data consist only of observations at 5 occasions, it was possible, after dividing the data into two groups, to fit plausible models in continuous time. Measurement error and alternative longitudinal designs are discussed, and some possible educational implications are noted.

Estimating Personality Impression from Speech Record Using Hidden Markov Models

2015 ◽  
Vol 135 (12) ◽  
pp. 1517-1523 ◽  
Author(s):  
Yicheng Jin ◽  
Takuto Sakuma ◽  
Shohei Kato ◽  
Tsutomu Kunitachi
Keyword(s):  
Hidden Markov Models ◽  
Markov Models ◽  
Hidden Markov
Sign in / Sign up

Export Citation Format

Share Document