Correlation and Causation
By Brian C. Mustain. 1937
In social science, we can rarely perform experiments rigorous enough to establish causality. Ironically, it is in just such areas of study that people are prone to rely on correlation to prove causation.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Joris M. Mooij
- Sara Magliacane
- Jan Peters
- Bernhard Schölkopf
- Stephan Bongers
- Tom Claassen
- Patrick Forr'e
- Dominik Janzing
Citations in the corpus
- The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation (2021, unknown): “Introduced by Wright (1921) and generally indicated by R 2 , its original formulation quanti fi es how much the dependent variable is determined by the independent variables, in terms of proportion of variance.”
- Joint Causal Inference from Multiple Contexts (2016, methodology): “2 Structural Causal Models Directed Mixed Graphs form a convenient graphical representation for variables (labelled by the nodes) and their functional relations (expressed by the edges) in a Structural Causal Model (SCM) (Pearl, 2009), also known as a (non-parametric) Structural Equation Model (SEM)”
- Self-Supervised Representation Learning on Neural Network Weights for Model Characteristic Prediction (2021, methodology): “Using such targets, we solve a regression problem and measure the R(2) score [Wright, 1921].”
- Foundations of structural causal models with cycles and latent variables (2016, methodology): “SCMs arose out of certain causal models that were first introduced in genetics [74], econometrics [22], electrical engineering [36, 37] and the social sciences [11, 21].”