Nideconv

Nideconv is a package that allows you to perform impulse response shape fitting on time series data, in order to estimate event-related signals.

Example use cases are fMRI and pupil size analysis. The package performs the linear least squares analysis using numpy.linalg as a backend, but can switch between different backends, such as statsmodels (which is not yet implemented). For very collinear design matrices ridge regression is implemented through the sklearn RidgeCV function.

It is possible to add covariates to the events to estimate not just the impulse response function, but also correlation timecourses with secondary variables. Furthermore, one can add the duration each event should have in the designmatrix, for designs in which the durations of the events vary.

In neuroscience, the inspection of the event-related signals such as those estimated by nideconv is essential for a thorough understanding of one's data. Researchers may overlook essential patterns in their data when blindly running GLM analyses without looking at the impulse response shapes.

Installation

Currently, nideconv can be installed using the GitHub repository:

Make Conda environment (optional but highly recomended, especially for Windows)

I highly recommend to first make a dedicated Anaconda/Miniconda-environment:

conda create --name nideconv

Then activate that environment

conda activate nideconv

Install nideconv

pip install git+https://github.com/VU-Cog-Sci/nideconv

Note: Due to the dependency on pystan for Bayesian analyses, which is currently not supported for Windows on Python versions >= 3.8.16, you will be able to install nideconv on Windows with Python versions > 3.8.16 but won't be able to use the Bayesian analysis functionality on higher Python versions (it will throw an error).

Documentation

The latest documentation can be found on http://nideconv.readthedocs.io/en/latest/