A Tale of Twin Dependence: A New Multivariate Regression Model and an FGLS Estimator for Analyzing Outcomes With Network Dependence

In this article, I present a new multivariate regression model for analyzing outcomes with network dependence. The model is capable to account for two types of outcome dependence including the mean dependence that allows the outcome to depend on selected features of a known dependence network and the error dependence that allows the outcome to be additionally correlated based on patterned connections in the dependence network (e.g., according to whether the ties are asymmetric, mutual, or triadic). For example, when predicting a group of students’ smoking status, the outcome can depend on the students’ positions in their friendship network and also be correlated among friends. I show that analyses ignoring the mean dependence can lead to severe bias in the estimated coefficients while analyses ignoring the error dependence can lead to inefficient inferences and failures in recognizing unmeasured social processes. I compare the new model with related models such as multilevel models, spatial regression models, and exponential random graph models and show their connections and differences. I propose a two-step, feasible generalized least squares estimator to estimate the model that is computationally fast and robust. Simulations show the validity of the new model (and the estimator) while four empirical examples demonstrate its versatility. Associated R package “fglsnet” is available for public use.

Diagnostic methods for uncovering outcome dependent visit processes

Summary With the advent of electronic health records, information collected in the course of regular health care is increasingly being used for clinical research. The hope is that the wealth of clinical data and the realistic setting (compared with information derived from highly controlled experiments like randomized trials) will aid in the investigation of determinants of disease and understanding of which treatments are effective in regular practice and for which patients. The availability of information in such databases is often driven by how a patient feels and may therefore be associated with the health outcomes being considered. We call this an outcome dependent visit process and recent work has shown that ignoring the outcome dependence can produce significant bias in the regression coefficients when fitting longitudinal data models. It is therefore important to have tools to recognize datasets exhibiting outcome dependence. We develop a score statistic to motivate the form of diagnostic test statistics, suggest a variety of approaches for diagnosing such situations, and evaluate their performance. Simple diagnostic tests achieve high power for diagnosing outcome dependent visit processes. This occurs when generalized estimating equations methods begin to be exhibit bias in estimating regression coefficients and before likelihood based methods are substantially biased.

A Theorem at the Core of Colliding Bias

Conditioning on a shared outcome of two variables can alter the association between these variables, possibly adding a bias component when estimating effects. In particular, if two causes are marginally independent, they might be dependent in strata of their common effect. Explanations of the phenomenon, however, do not explicitly state when dependence will be created and have been largely informal. We prove that two, marginally independent, causes will be dependent in a particular stratum of their shared outcome if and only if they modify each other’s effects, on a probability ratio scale, on that value of the outcome variable. Using our result, we also qualify the claim that such causes will “almost certainly” be dependent in at least one stratum of the outcome: dependence must be created in one stratum of a binary outcome, and independence can be maintained in every stratum of a trinary outcome.

Experimental test of nonlocal causality

Explaining observations in terms of causes and effects is central to empirical science. However, correlations between entangled quantum particles seem to defy such an explanation. This implies that some of the fundamental assumptions of causal explanations have to give way. We consider a relaxation of one of these assumptions, Bell’s local causality, by allowing outcome dependence: a direct causal influence between the outcomes of measurements of remote parties. We use interventional data from a photonic experiment to bound the strength of this causal influence in a two-party Bell scenario, and observational data from a Bell-type inequality test for the considered models. Our results demonstrate the incompatibility of quantum mechanics with a broad class of nonlocal causal models, which includes Bell-local models as a special case. Recovering a classical causal picture of quantum correlations thus requires an even more radical modification of our classical notion of cause and effect.

Modeling, Measuring, and Distinguishing Path Dependence, Outcome Dependence, and Outcome Independence

Path dependence is commonly used to describe processes where “history matters,” which encompasses many different kinds of temporal dynamics. This essay distinguishes path-, or equilibrium-, dependent processes where early conditions continue to matter, from outcome-dependent processes where recent history matters and from outcome-independent processes where history does not matter. Others have argued for a precise and restrictive definition of path dependence. We build on this and distinguish among different types of outcome-dependent processes when these conditions for path dependence are not fully satisfied.

A SELF-PRESENTATIONAL APPROACH TO TRANSMISSION OF GOOD AND BAD NEWS

People are reluctant to transmit bad news, a tendency named as the MUM effect. One explanation of this effect suggests that people do not want to construct negative impressions by being associated with bad news. In this study, transmission of good and bad news was examined from an impression management perspective. University students (N= 275) participated in a scenario study, with the valence of the news (good / bad) and outcome dependence on the recipient (high / low) as independent variables and transmission likelihood as dependent variable. Four variables, anticipated likeability, gratitude, perceived favor doing and ulterior motives, were assessed to form an ingratiation mediator. Results showed that participants were more likely to transmit good news than bad news. Also a significant interaction effect was obtained. In a high dependence condition participants were more likely to transmit good news and less likely to transmit bad news than participants in a low dependence condition. Moreover, the ingratiation construct significantly mediated the relationship investigated. In the second study (N = 74) similar findings were obtained except the interaction effect of dependence and news valence.