iiinova — where your digital self keeps evolving

Jan Failenschmid @janfailenschmid.bsky.social · 24/06/2026

4/4 I also want to thank my amazing supervisors and co-authors @leonievogelsmeier.bsky.social, Joris Mulder, and @joranjongerling.bsky.social. If you like anything in particular about this project, it was probably their suggestion or contribution.

120

Jan Failenschmid @janfailenschmid.bsky.social · 24/06/2026

3/4 So with this project, we aimed to improve the Bayesian estimation of these models and hopefully make them more accessible for empirical research. I will be at the ISBA next week to present this project. If you are there as well, I'm always happy to chat!

110

Jan Failenschmid @janfailenschmid.bsky.social · 24/06/2026

2/4 When we came across Gaussian Process State-Space Models we thought that they were a really exciting method for learning (multivariate) nonlinear dynamic relationships directly from time series data. However, we really struggled to estimate these models using the available tools.

100

Jan Failenschmid @janfailenschmid.bsky.social · 24/06/2026

1/4 In yet another act of shameless self-promotion, I want to share our new preprint: "Learning Nonlinear Dynamics: Improving the Estimation Efficiency and Reliability of Gaussian Process State-Space Models." Preprint: arxiv.org/abs/2606.24691 Any feedback or suggestions are very welcome!

arxiv.org

Learning Nonlinear Dynamics: Improving the Estimation Efficiency and Reliability of Gaussian Process State-Space Models

Understanding dynamic systems is a central goal in many scientific disciplines. State-space models provide a general framework for studying latent dynamic systems based on indirect observations. Howev...

183

Reposted by Jan Failenschmid

Leonie Vogelsmeier @leonievogelsmeier.bsky.social · 15/01/2026

📢 Call for Papers: We’re excited to announce an upcoming Psychometrika Special Issue on Data Intensive Methods in Psychometrics (think of using many datasets for methodological development), guest edited by @klint.bsky.social, @kyliegorney.bsky.social, @jmbh.bsky.social, Ben Domingue, and me.

psychometricsociety.org

055

Jan Failenschmid @janfailenschmid.bsky.social · 05/12/2025

I see, thanks for the explanation :)

100

Jan Failenschmid @janfailenschmid.bsky.social · 03/12/2025

Very good point. If you have time to explain it, I would be interested to learn how you got the intervals for each peak?

100

Reposted by Jan Failenschmid

Björn Siepe @bsiepe.bsky.social · 22/10/2025

We built the openESM database: ▶️60 openly available experience sampling datasets (16K+ participants, 740K+ obs.) in one place ▶️Harmonized (meta-)data, fully open-source software ▶️Filter & search all data, simply download via R/Python Find out more: 🌐 openesmdata.org 📝 doi.org/10.31234/osf...

14285147

Reposted by Jan Failenschmid

Laura Bringmann 🟥 @bringmannlaura.bsky.social · 26/06/2025

Want to learn about dynamic modeling for daily diary, experience sampling, ecological momentary assessment data? 😎 Register for our online course ‘Modeling the dynamics of intensive longitudindal data’ which starts in October 2025! 🤩 utrechtsummerschool.nl/courses/data...

utrechtsummerschool.nl

Modelling the Dynamics of Intensive Longitudinal Data (e-learning) 2025 | Utrecht Summer School

This online course covers how time series models can be used to model the dynamics of intensive longitudinal data (ILD).

12612

Jan Failenschmid @janfailenschmid.bsky.social · 02/06/2025

I’m really excited to share that our first article has been published in the Br. J. Math. Stat. Psychol. doi.org/10.1111/bmsp... In this paper, we evaluate and compare different non-parametric approaches for modeling non-linearity in psychological intensive longitudinal data.

doi.org

British Journal of Mathematical and Statistical Psychology | Wiley Online Library

Psychological concepts are increasingly understood as complex dynamic systems that change over time. To study these complex systems, researchers are increasingly gathering intensive longitudinal data...

143

Jan Failenschmid @janfailenschmid.bsky.social · 02/06/2025

Thank you very much and thank you for all your invaluable input, guidance, and support throughout this project.

010

Jan Failenschmid @janfailenschmid.bsky.social · 14/02/2025

Then I am looking forward to reading more about your analysis in the future. Do you think using a Bayesian model with priors on the number and location of the discontinuity points could be interesting for your data?

010

Jan Failenschmid @janfailenschmid.bsky.social · 14/02/2025

Did you by chance publish some more details on your data and analysis somewhere? I think it would be really interesting to see more on the robustness of this figure.

110

Jan Failenschmid @janfailenschmid.bsky.social · 13/02/2025

Thanks for the feedback, I finally got around to increase the thickness of the fitted curves. It does look better that way.

010

Reposted by Jan Failenschmid

Dariia Mykhailyshyna @handle.invalid · 07/02/2025

❗️Our next workshop will be on February 13th, 6 pm CET on Gaussian Process Regression in R and Stan by @janfailenschmid.bsky.social! Register or sponsor a student by donating to support Ukraine! Details: bit.ly/3wBeY4S Please share! #AcademicSky #EconSky #RStats

1128

Reposted by Jan Failenschmid

Dariia Mykhailyshyna @handle.invalid · 28/12/2024

On February 13 we will have a workshop on Modeling Non-Linear Relationships: An Introduction to Gaussian Process Regression in R and Stan by @janfailenschmid.bsky.social More info: bit.ly/4iQSzI7 Please share! #RStats #EconSky #AcademicSky

131

Jan Failenschmid @janfailenschmid.bsky.social · 24/10/2024

I am very grateful to my incredible supervisors and co-authors, @leonievogelsmeier.bsky.social , Joris Mulder, and @joranjongerling.bsky.social, for their invaluable contributions, guidance, and endless support throughout this project. 3/3

120

Jan Failenschmid @janfailenschmid.bsky.social · 24/10/2024

In this project, we review three non-parametric and non-linear regression techniques - local polynomial regression, Gaussian processes, and generalized additive models - within the context of intensive longitudinal data and compare how well these methods can recover psychological processes. 2/3

110

Jan Failenschmid @janfailenschmid.bsky.social · 24/10/2024

New Preprint! I’m excited to share our preprint titled: "Modeling Non-Linear Psychological Processes: Reviewing and Evaluating Non-Parametric Approaches and Their Applicability to Intensive Longitudinal Data." Check out the full preprint here: osf.io/preprints/ps... Any feedback is welcome! 1/3

osf.io

OSF

2143