DerivKit: stable numerical derivatives bridging Fisher forecasts and MCMC

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Šarčević, Nikolina, van der Wild, Matthijs, Trendafilova, Cynthia
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910015654199296
author Šarčević, Nikolina
van der Wild, Matthijs
Trendafilova, Cynthia
author_facet Šarčević, Nikolina
van der Wild, Matthijs
Trendafilova, Cynthia
contents DerivKit is a Python package for derivative-based statistical inference. It implements stable numerical differentiation and derivative assembly utilities for Fisher-matrix forecasting and higher-order likelihood approximations in scientific applications, supporting scalar- and vector-valued models including black-box or tabulated functions where automatic differentiation is impractical or unavailable. These derivatives are used to construct Fisher forecasts, Fisher bias estimates, and non-Gaussian likelihood expansions based on the Derivative Approximation for Likelihoods (DALI). By extending derivative-based inference beyond the Gaussian approximation, DerivKit forms a practical bridge between fast Fisher forecasts and more computationally intensive sampling-based methods such as Markov chain Monte Carlo (MCMC).
format Preprint
id arxiv_https___arxiv_org_abs_2602_08078
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DerivKit: stable numerical derivatives bridging Fisher forecasts and MCMC
Šarčević, Nikolina
van der Wild, Matthijs
Trendafilova, Cynthia
Instrumentation and Methods for Astrophysics
Cosmology and Nongalactic Astrophysics
Data Analysis, Statistics and Probability
DerivKit is a Python package for derivative-based statistical inference. It implements stable numerical differentiation and derivative assembly utilities for Fisher-matrix forecasting and higher-order likelihood approximations in scientific applications, supporting scalar- and vector-valued models including black-box or tabulated functions where automatic differentiation is impractical or unavailable. These derivatives are used to construct Fisher forecasts, Fisher bias estimates, and non-Gaussian likelihood expansions based on the Derivative Approximation for Likelihoods (DALI). By extending derivative-based inference beyond the Gaussian approximation, DerivKit forms a practical bridge between fast Fisher forecasts and more computationally intensive sampling-based methods such as Markov chain Monte Carlo (MCMC).
title DerivKit: stable numerical derivatives bridging Fisher forecasts and MCMC
topic Instrumentation and Methods for Astrophysics
Cosmology and Nongalactic Astrophysics
Data Analysis, Statistics and Probability
url https://arxiv.org/abs/2602.08078