This paper examines Patrick Suppes’ probabilistic theory of causality understood as a theory of causal inference, and draws some lessons for empirical economics and contemporary debates in the foundations of econometrics. It argues that a standard method of empirical economics, multiple regression, is inadequate for most but the simplest applications, that the Bayes’ nets approach, which can be understood as a generalisation of Suppes’ theory, constitutes a considerable improvement but is still subject to important limitations, and that the currently fashionable (...) ‘design-based approach’ suffers from the same flaws Suppes anticipated a long time ago. It then sketches an alternative in response, one that differs drastically from the formalisms Suppes endorsed but is consistent with his pragmatic general take on science. (shrink)

In the increasing body of metatheoretical literature on "causal mechanisms," definitions of "mechanism" proliferate, and these increasingly divergent definitions reproduce older theoretical and methodological oppositions. The reason for this proliferation is the incompatibility of the various metatheoretical expectations directed to them: (1) to serve as an alternative to the scientific theory of individual behavior (for some social theorists, most notably Jon Elster); (2) to provide solutions for causal inference problems in the quantitative social sciences, in social history, and in the (...) (3) qualitative research context; and (4) to serve as an alternative for narratives (Charles Tilly). Mechanisms can do (1) only as under-specified law-like regularities, deliver (2) as robust generative processes represented by models, and accomplish (3) as fragile generative processes (stories), but these are not all compatible. (edited). (shrink)

We further develop the mathematical theory of causal interventions, extending earlier results of Korb, Twardy, Handfield, & Oppy, (2005) and Spirtes, Glymour, Scheines (2000). Some of the skepticism surrounding causal discovery has concerned the fact that using only observational data can radically underdetermine the best explanatory causal model, with the true causal model appearing inferior to a simpler, faithful model (cf. Cartwright, (2001). Our results show that experimental data, together with some plausible assumptions, can reduce the space of viable explanatory (...) causal models to one. (shrink)

I consider three aspects in which machine learning and philosophy of science can illuminate each other: methodology, inductive simplicity and theoretical terms. I examine the relations between the two subjects and conclude by claiming these relations to be very close.

Bayesian reasoning has been applied formally to statistical inference, machine learning and analysing scientific method. Here I apply it informally to more common forms of inference, namely natural language arguments. I analyse a variety of traditional fallacies, deductive, inductive and causal, and find more merit in them than is generally acknowledged. Bayesian principles provide a framework for understanding ordinary arguments which is well worth developing.

There have been many efforts to infer causation from association byusing statistical models. Algorithms for automating this processare a more recent innovation. In Humphreys and Freedman[(1996) British Journal for the Philosophy of Science 47, 113–123] we showed that one such approach, by Spirtes et al., was fatally flawed. Here we put our arguments in a broader context and reply to Korb and Wallace [(1997) British Journal for thePhilosophy of Science 48, 543–553] and to Spirtes et al.[(1997) British Journal for the (...) Philosophy of Science 48, 555–568]. Their arguments leave our position unchanged: claims to have developed a rigorous engine for inferring causation from association are premature at best, the theorems have no implications for samples of any realistic size, and the examples used to illustrate the algorithms are indicative of failure rather than success. The gap between association and causation has yet to be bridged. (shrink)

The theory of fast and frugal heuristics, developed in a new book called Simple Heuristics that make Us Smart (Gigerenzer, Todd, and the ABC Research Group, in press), includes two requirements for rational decision making. One is that decision rules are bounded in their rationality –- that rules are frugal in what they take into account, and therefore fast in their operation. The second is that the rules are ecologically adapted to the environment, which means that they `fit to reality.' (...) The main purpose of this article is to apply these ideas to learning rules–-methods for constructing, selecting, or evaluating competing hypotheses in science, and to the methodology of machine learning, of which connectionist learning is a special case. The bad news is that ecological validity is particularly difficult to implement and difficult to understand. The good news is that it builds an important bridge from normative psychology and machine learning to recent work in the philosophy of science, which considers predictive accuracy to be a primary goal of science. (shrink)

What Do the Data Tell Us? Justification of scientific theories is a three-place relation between data, theories, and background knowledge. Though this should be a commonplace, many methodologies in science neglect it. The article will elucidate the significance and function of our background knowledge in epistemic justification and their consequences for different scientific methodologies. It is argued that there is no simple and at the same time acceptable statistical algorithm that justifies a given theory merely on the basis of certain (...) data. And even if we think to know the probability of a theory, that does not decide whether we should accept it or not. (shrink)