Implausible Assumptions Research Articles

Detecting spurious effects in cross-lagged panel models: Triangulation is not a valid test

The cross-lagged panel model (CLPM) is an analytic technique used to examine the reciprocal causal effects of two or more variables assessed on two or more occasions. Although widely used, the CLPM has been criticized for relying on implausible assumptions, the violation of which can often lead to biased estimates of causal effects. Recently, a triangulation method has been proposed to identify spurious effects in simple CLPM analyses (e.g., Sorjonen, Melin, & Melin, 2024). We use simulations and a discussion of the formulas underlying regression coefficients to show that this method does not provide a valid indicator of spuriousness. This method identifies true causal effects as spurious in realistic situations and should not be used to diagnose whether a causal effect estimated from the CLPM is spurious or not. There are clear reasons to doubt causal estimates from the CLPM, but the results of the triangulation method do not add information about whether such estimates are spurious.

Endogenous Preference for Nonmarket Goods in Carbon Abatement Decisions

Long-term decisions, such as decisions about carbon abatement, are usually based on the implausible assumption of constant social preference. This paper focuses on a specific case of market and nonmarket goods, and it investigates the optimal climate policy when social preference for them is also changed by climate policy in a seminal climate-economy model: the Dynamic Integrated Model of Climate and the Economy (DICE) model with a single objective to maximize market goods consumption. We amend DICE to study an optimization problem that trades off the values of market and nonmarket goods, considering endogenous preferences. The relative price of non-market goods grows over time because of increases in both relative scarcity and the appreciation that represents the preference of it. As nonmarket goods are increasingly valuable relative to market ones, climate damages on nonmarket goods are increasingly expensive. The social cost of carbon, which aggregated climate impact on both goods, is consequently higher. Because abatement decision affects the valuation of nonmarket goods in the utility function, unlike previous climate-economy models, we solve the model iteratively by taking the obtained abatement rates from the last run as inputs in the current run. The results in baseline calibration advocate a more stringent climate policy, where endogenous social preference to climate policy raises the social cost of carbon further by roughly 12%–18% this century. Moreover, neglecting changing social preference leads to an underestimate of nonmarket goods damages by 15%. Climate policy is self-reinforced if it favors the more expensive consumption good. The proposed method can be applied to other long-term problems with endogenous preferences. Funding: H. Liao appreciates the funding from the National Natural Science Foundation of China [Grants 71925008, 72293603, and 72488101].

Word Frequency and Predictability Dissociate in Naturalistic Reading.

Many studies of human language processing have shown that readers slow down at less frequent or less predictable words, but there is debate about whether frequency and predictability effects reflect separable cognitive phenomena: are cognitive operations that retrieve words from the mental lexicon based on sensory cues distinct from those that predict upcoming words based on context? Previous evidence for a frequency-predictability dissociation is mostly based on small samples (both for estimating predictability and frequency and for testing their effects on human behavior), artificial materials (e.g., isolated constructed sentences), and implausible modeling assumptions (discrete-time dynamics, linearity, additivity, constant variance, and invariance over time), which raises the question: do frequency and predictability dissociate in ordinary language comprehension, such as story reading? This study leverages recent progress in open data and computational modeling to address this question at scale. A large collection of naturalistic reading data (six datasets, >2.2 M datapoints) is analyzed using nonlinear continuous-time regression, and frequency and predictability are estimated using statistical language models trained on more data than is currently typical in psycholinguistics. Despite the use of naturalistic data, strong predictability estimates, and flexible regression models, results converge with earlier experimental studies in supporting dissociable and additive frequency and predictability effects.

Precedent and rest stop convergence in reflective equilibrium

The method of reflective equilibrium is typically characterized as a process of two kinds of adjustments: hold fixed one’s current set of commitments/intuitions and adjust rules/principles to account for them, then hold fixed those rules while making adjustments to one’s set of commitments. Repeat until no further adjustments are required. Such characterizations ignore the role of precedent, i.e., information about the commitments and rules of others and how those might serve as guides in one’s own process of deliberation. In this paper we develop a model that narrows in on a local part of a reflective equilibrium process in which a group of peer agents start with identical lists of commitments and work towards finding the rule that captures that set. If successful, we say they reach rest stop interpersonal convergence. The trouble is that, without precedent, reaching this convergence rests on implausible deterministic assumptions. We analyze how far the inclusion of a kind of soft precedent goes towards aiding the group in reaching said convergence when those assumptions are relaxed.

Moderate realist ideology critique

AbstractRealist ideology critique (RIC) is a strand of political realism recently developed in response to concerns that realism is biased toward the status quo. RIC aims to debunk an individual's belief that a social institution is legitimate by revealing that the belief is caused by that very same institution. Despite its growing prominence, RIC has received little critical attention. In this article, I buck this trend. First, I improve on contemporary accounts of RIC by clarifying its status and the role of motivated reasoning. Second, I show that realist ideology critics face a dilemma: either their account makes deeply implausible epistemological assumptions, or they temper its epistemology at the expense of rendering it toothless. I argue for each horn in turn before revealing the dilemma to be a false one by making a novel distinction between varying strengths of RIC based on their underlying epistemological assumptions. I propose Moderate RIC as a solution: upon discovering that one reason for your belief that a social institution is legitimate is likely malignantly epistemically circular, the belief should undergo further epistemic testing. I respond to three potential objections and suggest that Moderate RIC would make a fruitful addition to political theorists' methodological toolkit.

Logical Inconsistencies in the Health Years in Total and Equal Value of Life-Years Gained

ObjectivesThis study aimed to assess whether recently proposed alternatives to the quality-adjusted life-year (QALY), intended to address concerns about discrimination, are suitable for informing resource allocation decisions. MethodsWe consider 2 alternatives to the QALY: the health years in total (HYT), recently proposed by Basu et al, and the equal value of life-years gained (evLYG), currently used by the Institute for Clinical and Economic Review. For completeness we also consider unweighted life-years (LYs). Using a hypothetical example comparing 3 mutually exclusive treatment options, we consider how calculations are performed under each approach and whether the resulting rankings are logically consistent. We also explore some further challenges that arise from the unique properties of the HYT approach. ResultsThe HYT and evLYG approaches can result in logical inconsistencies that do not arise under the QALY or LY approaches. HYT can violate the independence of irrelevant alternatives axiom, whereas the evLYG can produce an unstable ranking of treatment options. HYT have additional issues, including an implausible assumption that the utilities associated with health-related quality of life and LYs are “separable,” and a consideration of “counterfactual” health-related quality of life for patients who are dead. ConclusionsThe HYT and evLYG approaches can result in logically inconsistent decisions. We recommend that decision makers avoid these approaches and that the logical consistency of any approaches proposed in future be thoroughly explored before considering their use in practice.

Reason monolithism: A Darwinian dilemma for “relaxed” realism

AbstractStreet formulated a Darwinian Dilemma for realist theories of value. Much criticism of her formulation of the dilemma targets the second horn, posed by the scientifically implausible assumption of a tracking relation between our attitudes and evaluative truth. This paper shows how a recent wave of metaethical realism, most prominently defended by Scanlon, succeeds without a tracking relation and thus avoids the Darwinian Dilemma in Street's formulation. However, Scanlon's approach, which builds on the concept of a reason relation and defends a metaphysically pluralist, domain‐specific conception of truth, runs into another version of the Darwinian Dilemma. The problem is not that Scanlon's realism assumes a tracking relation but that it assumes what I call reason monolithism – the idea that there is one possible expression of the faculty of reason and that this cognitive faculty could not be otherwise, which is scientifically implausible on similar grounds.

Facing up to the problem of intentionality

AbstractWe distinguish between different problems of “aboutness”: the “hard” problem of explaining the everyday phenomenon of intentionality and three less challenging “easy” sets of problems concerning the posits of folk psychology, the notions of representation invoked in the mind‐brain sciences, and the intensionality (with an “s”) of mental language. The problem of intentionality is especially hard in that, as is the case with the hard problem of phenomenal consciousness, there is no clear path to a solution using current methods. We argue that naturalistic theories of mental representation do not address the hard problem—either they are only intended to address the easy problems, or the claims they make help address the problem of intentionality only under undefended and prima facie implausible assumptions to the effect that the hard problem reduces to some combination of the easy problems. We offer a positive account of what would be required to properly face up to the problem of intentionality.

Alternatives to traditional outcome modelling approaches in applied linguistics: A primer on propensity score matching

Observational studies and quasi-experiments are commonplace in applied linguistics. Causal inference based on such a design hinges on different assumptions, one of which is commonly referred to as the assumption of no unmeasured confounding. To address this assumption, researchers typically consider different confounders using regression adjustment methods (i.e., entering the confounders into the model as covariates). Although these methods are theoretically sound procedures, they are not without criticism, for instance, regarding potentially extrapolated effect estimates or implausible linear functional form assumptions. In this article, we introduce an alternative statistical method, known as propensity score matching (PSM), which can help address the limitations of traditional regression adjustment methods. We use an example case study, where we first discuss the challenges applied linguists face when they intend to investigate causal effects. We illustrate how PSM can be used with applied linguistics data and examine theoretical and practical issues around each analytical step, including electing relevant covariates, estimating the propensity score, performing matching, assessing the matches, and estimating the treatment effects. Our overall goal is to provide applied linguists with an alternative method (PSM), which might be beneficial and reasonably applied in several practical situations.

The Transcendental Phenomenological Argument against Eternalism

Abstract In this paper, we argue against eternalism on the basis of certain phenomenological considerations regarding our experiential life in a relatively novel way. Contrary to well-known phenomenological arguments that attempt to refute tenseless theories of time, our argument that we call the Transcendental Phenomenological Argument against Eternalism is against both tenseless and tensed versions of eternalism. The argument is based on the fact that one experiences a phenomenological succession of experiences, and it shows that perdurantist forms of eternalism have to either deny this fact or should embrace ad hoc and metaphysically implausible assumptions about the nature of the mind. As we argue, neither of these options seems to be too promising.

Bounding the average causal effect in Mendelian randomisation studies with multiple proposed instruments: An application to prenatal alcohol exposure and attention deficit hyperactivity disorder.

As large-scale observational data become more available, caution regarding causal assumptions remains critically important. This may be especially true for Mendelian randomisation (MR), an increasingly popular approach. Point estimation in MR usually requires strong, often implausible homogeneity assumptions beyond the core instrumental conditions. Bounding, which does not require homogeneity assumptions, is infrequently applied in MR. We aimed to demonstrate computing nonparametric bounds for the causal risk difference derived from multiple proposed instruments in an MR study where effect heterogeneity is expected. Using data from the Norwegian Mother, Father and Child Cohort Study (n=2056) and Avon Longitudinal Study of Parents and Children (n=6216) to study the average causal effect of maternal pregnancy alcohol use on offspring attention deficit hyperactivity disorder symptoms, we proposed 11 maternal SNPs as instruments. We computed bounds assuming subsets of SNPs were jointly valid instruments, for all combinations of SNPs where the MR model was not falsified. The MR assumptions were violated for all sets with more than 4 SNPs in one cohort and for all sets with more than 2 SNPs in the other. Bounds assuming one SNP was an individually valid instrument barely improved on assumption-free bounds. Bounds tightened as more SNPs were assumed to be jointly valid instruments, and occasionally identified directions of effect, though bounds from different sets varied. Our results suggest that, when proposing multiple instruments, bounds can contextualise plausible magnitudes and directions of effects. Computing bounds over multiple assumption sets, particularly in large, high-dimensional data, offers a means of triangulating results across different potential sources of bias within a study and may help researchers to better evaluate and emphasise which estimates are compatible with the most plausible assumptions for their specific setting.

A causal roadmap for generating high-quality real-world evidence.

Increasing emphasis on the use of real-world evidence (RWE) to support clinical policy and regulatory decision-making has led to a proliferation of guidance, advice, and frameworks from regulatory agencies, academia, professional societies, and industry. A broad spectrum of studies use real-world data (RWD) to produce RWE, ranging from randomized trials with outcomes assessed using RWD to fully observational studies. Yet, many proposals for generating RWE lack sufficient detail, and many analyses of RWD suffer from implausible assumptions, other methodological flaws, or inappropriate interpretations. The Causal Roadmap is an explicit, itemized, iterative process that guides investigators to prespecify study design and analysis plans; it addresses a wide range of guidance within a single framework. By supporting the transparent evaluation of causal assumptions and facilitating objective comparisons of design and analysis choices based on prespecified criteria, the Roadmap can help investigators to evaluate the quality of evidence that a given study is likely to produce, specify a study to generate high-quality RWE, and communicate effectively with regulatory agencies and other stakeholders. This paper aims to disseminate and extend the Causal Roadmap framework for use by clinical and translational researchers; three companion papers demonstrate applications of the Causal Roadmap for specific use cases.

Attention-Based Cross-Modal CNN Using Non-Disassembled Files for Malware Classification

The role of malware classification is crucial in addressing the explosive increase in malware variants. By classifying malware instances into malware families, malware analysts can apply appropriate techniques and tools to handle malware variants in each family. Using high-level representations of malware, such as disassembled codes, yields meaningful classification performance. However, malware classification based on disassembled codes depends on the practically implausible assumption that every malware is correctly reversed by disassemblers. Unfortunately, sophisticated malware, which has anti-disassembly capabilities, seeks to confuse disassemblers, yielding incorrectly disassembled codes. In this study, we focus on malware family classification, which requires no disassembly, and propose a new CNN-based malware classification model using non-disassembled malware files (i.e., binary files). Our model associates two modalities: “malware images” and “structural entropies,” which are converted and extracted from binary files. Both modalities have different granularities of bytes and chunks that complement each other. The model adopts a cross-modal attention mechanism to combine the features of the two modalities by moderating their expressive limitations. We validate our model using three popular datasets from the Kaggle Microsoft Malware Classification, Malimg, and BODMAS datasets. The experimental results show that our model identifies malware families with a higher degree of accuracy than previous methods and does not require the burden of disassembling.

Four-Parameter Guessing Model and Related Item Response Models

Guessing effects frequently occur in testing data in educational or psychological applications. Different item response models have been proposed to handle guessing effects in dichotomous test items. However, it has been pointed out in the literature that the often employed three-parameter logistic model poses implausible assumptions regarding the guessing process. The four-parameter guessing model has been proposed as an alternative to circumvent these conceptual issues. In this article, the four-parameter guessing model is compared with alternative item response models for handling guessing effects through a simulation study and an empirical example. It turns out that model selection for item response models should be rather based on the AIC than the BIC. However, the RMSD item fit statistic used with typical cutoff values was found to be ineffective in detecting misspecified item response models. Furthermore, sufficiently large sample sizes are required for sufficiently precise item parameter estimation. Moreover, it is argued that the criterion of the statistical model fit should not be the sole criterion of model choice. The item response model used in operational practice should be valid with respect to the meaning of the ability variable and the underlying model assumptions. In this sense, the four-parameter guessing model could be the model of choice in educational large-scale assessment studies.

The collapse of logical contextualism

ABSTRACT The most serious objection to Beall and Restall’s case-based logical pluralism is the so-called collapse argument. According to the collapse argument, logical pluralism is not genuinely pluralistic and collapses into a single privileged relation of logical consequence. In response, Caret offered an account of logical contextualism that supposedly maintains the merits of Beall and Restall’s case-based logical pluralism while circumventing the collapse argument. In this paper, I first point out a gap in the collapse argument in that it does not affect pluralists who assume that there is no uniquely strongest admissible consequence relation. This gap, then, is closed by an improved version of the collapse argument, which makes use of Eklund’s classification of types of pluralism. Finally, I discuss how Caret’s logical contextualism cannot hold up against this new formulation of the collapse argument unless implausible assumptions about the contents of normative operators are made. Moreover, by reformulating the collapse argument, I determine the extent to which logical monism and pluralism oppose each other, paving the way for fruitful future discussion.

Nonparametric, Stochastic Frontier Models with Multiple Inputs and Outputs

Stochastic frontier models along the lines of Aigner et al. are widely used to benchmark firms’ performances in terms of efficiency. The models are typically fully parametric, with functional form specifications for the frontier as well as both the noise and the inefficiency processes. Studies such as Kumbhakar et al. have attempted to relax some of the restrictions in parametric models, but so far all such approaches are limited to a univariate response variable. Some (e.g., Simar and Zelenyuk; Kuosmanen and Johnson) have proposed nonparametric estimation of directional distance functions to handle multiple inputs and outputs, raising issues of endogeneity that are either ignored or addressed by imposing restrictive and implausible assumptions. This article extends nonparametric methods developed by Simar et al. and Hafner et al. to allow multiple inputs and outputs in an almost fully nonparametric framework while avoiding endogeneity problems. We discuss properties of the resulting estimators, and examine their finite-sample performance through Monte Carlo experiments. Practical implementation of the method is illustrated using data on U.S. commercial banks.

The Econ within or the Econ above? On the plausibility of preference purification

AbstractScholars disagree about the plausibility of preference purification. Some see it as a familiar phenomenon. Others denounce it as conceptually incoherent, postulating that it relies on the psychologically implausible assumption of an inner rational agent. I argue that different notions of rationality can be leveraged to advance the debate: procedural rationality and structural rationality. I explicate how structural rationality, in contrast to procedural rationality, allows us to offer an account of the guiding idea behind preference purification that avoids inner rational agents. Afterward, I address two pressing challenges against preference purification that emerge under the structural rationality account.

Relativism and Two Kinds of Branching Time

AbstractThis essay examines the case for relativism about future contingents in light of a distinction between two ways of interpreting the ‘branching time’ framework. The first step of the relativist argument is to argue for the ‘Non‐Determination Thesis’, the view that there is no unique actual future. The second step is to argue from the Non‐Determination Thesis to relativism. I show that first step of this argument fails. But despite that result, the second step is still of interest, because one might hold the Non‐Determination Thesis on alternative grounds. I then argue that whether the second step of the argument succeeds depends on how the branches in question are interpreted. If the branches are ersatz possible worlds, then the argument for relativism might go through. But if the branches are concrete parts of a ‘branching multiverse’, then the argument for relativism turns out to make implausible assumptions about the nature of personal identity over time.

Sharp Sensitivity Analysis for Inverse Propensity Weighting via Quantile Balancing

Inverse propensity weighting (IPW) is a popular method for estimating treatment effects from observational data. However, its correctness relies on the untestable (and frequently implausible) assumption that all confounders have been measured. This article introduces a robust sensitivity analysis for IPW that estimates the range of treatment effects compatible with a given amount of unobserved confounding. The estimated range converges to the narrowest possible interval (under the given assumptions) that must contain the true treatment effect. Our proposal is a refinement of the influential sensitivity analysis by Zhao, Small, and Bhattacharya, which we show gives bounds that are too wide even asymptotically. This analysis is based on new partial identification results for Tan’s marginal sensitivity model. Supplementary materials for this article are available online.

A General Method for Deriving Tight Symbolic Bounds on Causal Effects

A causal query will commonly not be identifiable from observed data, in which case no estimator of the query can be contrived without further assumptions or measured variables, regardless of the amount or precision of the measurements of observed variables. However, it may still be possible to derive symbolic bounds on the query in terms of the distribution of observed variables. Bounds, numeric or symbolic, can often be more valuable than a statistical estimator derived under implausible assumptions. Symbolic bounds, however, provide a measure of uncertainty and information loss due to the lack of an identifiable estimand even in the absence of data. We develop and describe a general approach for computation of symbolic bounds and characterize a class of settings in which our method is guaranteed to provide tight valid bounds. This expands the known settings in which tight causal bounds are solutions to linear programs. We also prove that our method can provide valid and possibly informative symbolic bounds that are not guaranteed to be tight in a larger class of problems. We illustrate the use and interpretation of our algorithms in three examples in which we derive novel symbolic bounds.