Model estimation and evaluation

This is the landing page for topics related to estimation and evaluation of models for your intensive longitudinal data (ILD).

There are various techniques that can be used to estimate ILD models. An important distinction between these techniques is whether they are based on N=1 or N>1 data, that is, whether you want to analyze data of a single case or multiple cases (i.e., persons, dyads, or other entities). There are also various techniques to evaluate the model you have estimated; how to evaluate a model depends very much on what your research question and goal are.

Below we have specified multiple articles where you can read more about estimation of ILD models.

Think more about estimation

Model estimation techniques

Various techniques that you can use to estimate ILD models.

• Kalman filter
• State-space model

Model evaluation techniques

There are various techniques that you can use to evaluate ILD models.

• [White noise]
• [Cross-validation]

Factors that play a role in estimating and evaluating model

How to decide between the various ways to estimate and evaluate a model depends on various factors.

• N=1 versus N>1 analysis
• Analysis per research goal

---

Ellen L. Hamaker

2025-07-11