Causal models

Research group type

Research group

Core fields of research

Basic natural phenomena and mathematical thinking

Research areas

Statistics

Faculty

Faculty of Mathematics and Science

Department

Department of Mathematics and Statistics

Research group description

Causality is central for science and decision making. Causal relations are often visualized using graphs. We study causal effect identification. The methods and algorithms we have developed combine experimental and observational data sources and take into account misssing data and selection bias.

Selected publications

S. Tikka, J. Helske, J.Karvanen, Clustering and structural robustness in causal diagrams, Journal of Machine Learning Research, 24(195):1-32, 2023.
Abstract
Open access article
Code
T. Gorbach, X. de Luna, J. Karvanen, I. Waernbaum, Contrasting Identifying assumptions of average causal effects: Robustness and semiparametric efficiency, Journal of Machine Learning Research, 24(197):1-65, 2023.
Abstract
Open access article
Code
S. Tikka, A. Hyttinen, J. Karvanen, Causal effect identification from multiple incomplete data sources: A general search-based approach, Journal of Statistical Software, 99(5), 2021.
Open access article
R package dosearch
S. Tikka, A. Hyttinen, J. Karvanen, Identifying causal effects via context-specific independence relations, 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), http://papers.nips.cc/paper/8547-identifying-causal-effects-via-context…, 2019.
Open access PDF
Supplement
Poster
S. Tikka, J. Karvanen, Enhancing identification of causal effects by pruning, Journal of Machine Learning Research, 18(194):1-23, 2018.
Open access article
Preprint JYX
R package causaleffect by Santtu Tikka

Open source R-packages created by the research group

causaleffect	Deriving Expressions of Joint Interventional Distributions and Transport Formulas in Causal Models	Santtu Tikka santtuth@gmail.com
cfid	Identification of Counterfactual Queries in Causal Models	Santtu Tikka santtuth@gmail.com
dosearch	Causal Effect Identification from Multiple Incomplete Data Sources	Santtu Tikka santtuth@gmail.com
R6causal	R6 Class for Structural Causal Models	Juha Karvanen juha.t.karvanen@jyu.fi

Causal models

Table of contents

Research group description

Selected publications

Open source R-packages created by the research group

Research group

Group leaders

Juha Karvanen

Santtu Tikka

Members

Otto Tabell