Data from: Using hidden Markov models to improve quantifying physical activity in accelerometer data – a simulation study

Witowski V, Foraita R, Pitsiladis Y, Pigeot I, Wirsik N

Date Published: December 3, 2014

DOI: http://dx.doi.org/10.5061/dryad.tq0gt

 

Files in this package

Content in the Dryad Digital Repository is offered "as is." By downloading files, you agree to the Dryad Terms of Service. To the extent possible under law, the authors have waived all copyright and related or neighboring rights to this data. CC0 (opens a new window) Open Data (opens a new window)

Title Generated Accelerometer data for Simulation Study (Format RData)
Downloaded 36 times
Description Each R-File (Day1-Day1000) contains the list object "data" which contains the complete generated data for a day. The simulations were designed to reflect free living-environment accelerometer observations obtained for children. 1,000 different time series with the length of T=1,440 and an epoch length of 15 seconds were simulated, each representing 6 hours of counts per day. Counts per day were randomly generated using the negative binomial distribution (with parameters r=1 and p=0.0009, resulting in the lowest PA-level μ_1=111.11) and the Gaussian distribution (with the parameters μ_2=400, μ_3=600 and μ_4=900, with σ_2^2=σ_3^2=σ_4^2=10,000) around three or four pre-defined physical activity levels (depending on the random time series generated by a Markov chain).
Download Format RData.zip (5.120 Mb)
Download README.pdf (114.1 Kb)
Details View File Details
Title Generated Accelerometer data for Simulation Study (Format rds)
Downloaded 15 times
Description Each R-File (Day1-Day1000) contains the list object "data" which contains the complete generated data for a day. The simulations were designed to reflect free living-environment accelerometer observations obtained for children. 1,000 different time series with the length of T=1,440 and an epoch length of 15 seconds were simulated, each representing 6 hours of counts per day. Counts per day were randomly generated using the negative binomial distribution (with parameters r=1 and p=0.0009, resulting in the lowest PA-level μ_1=111.11) and the Gaussian distribution (with the parameters μ_2=400, μ_3=600 and μ_4=900, with σ_2^2=σ_3^2=σ_4^2=10,000) around three or four pre-defined physical activity levels (depending on the random time series generated by a Markov chain).
Download Format rds.zip (5.101 Mb)
Download README.rtf (58.25 Kb)
Details View File Details

When using this data, please cite the original publication:

Witowski V, Foraita R, Pitsiladis Y, Pigeot I, Wirsik N (2014) Using hidden Markov models to improve quantifying physical activity in accelerometer data – a simulation study. PLoS ONE 9(12): e114089. http://dx.doi.org/10.1371/journal.pone.0114089

Additionally, please cite the Dryad data package:

Witowski V, Foraita R, Pitsiladis Y, Pigeot I, Wirsik N (2014) Data from: Using hidden Markov models to improve quantifying physical activity in accelerometer data – a simulation study. Dryad Digital Repository. http://dx.doi.org/10.5061/dryad.tq0gt
Cite | Share
Download the data package citation in the following formats:
   RIS (compatible with EndNote, Reference Manager, ProCite, RefWorks)
   BibTex (compatible with BibDesk, LaTeX)

Search for data

Be part of Dryad

We encourage organizations to: