Direction Dependence Analysis in the Presence of Confounders: Applications to Linear Mediation Models Using Observational Data

2019 ◽  
Vol 55 (4) ◽  
pp. 495-515 ◽  
Author(s):  
Wolfgang Wiedermann ◽  
James Sebastian
Keyword(s):  
Observational Data ◽  
Dependence Analysis ◽  
Direction Dependence ◽  
Mediation Models

Direction-dependence analysis

2015 ◽  
Vol 39 (6) ◽  
pp. 570-580 ◽  
Author(s):  
Wolfgang Wiedermann ◽  
Alexander von Eye
Keyword(s):  
Real World ◽  
Empirical Evaluation ◽  
Data Sets ◽  
Dependence Analysis ◽  
Linear Regression Models ◽  
Real World Data ◽  
Variable Pair ◽  
Causal Search ◽  
Direction Dependence ◽  
New Guidelines

The concept of direction dependence has attracted growing attention due to its potential to help decide which of two competing linear regression models ( X → Y or Y → X) is more likely to reflect the correct causal flow. Several tests have been proposed to evaluate hypotheses compatible with direction dependence. In this issue, Thoemmes (2015) reports results of an empirical evaluation of direction-dependence tests using real-world data sets with known causal ordering and concludes that the tests (known to perform excellent in simulation studies) perform poorly in the real-world setting. The present article aims at answering the question how this is possible. First, we review potential conceptual issues associated with Thoemmes’ (2015) approach. We argue that direction dependence is best conceptualized as a confirmatory approach to test focused directional theories. Thoemmes’ (2015) evaluation is based on an exploratory use of direction dependence. It implicitly follows the tradition of causal search algorithms. Second, we discuss potential statistical issues associated with Thoemmes’ (2015) selection schemes used to decide whether a variable pair is suitable for direction-dependence analysis. Based on these issues, new tests of direction dependence as well as new guidelines for confirmatory direction-dependence analysis are proposed. An empirical example is given to illustrate the application of these guidelines.

scholarly journals Vlbi research in the ”Radioastron” project: structure of the H2O maser in NGC 2071 IRS 1

2021 ◽  
Author(s):  
M. A. Shchurov ◽  
◽  
I. E. Valtts ◽  
N. N. Shakhvorostova ◽  
◽  
...  
Keyword(s):  
Spatial Structure ◽  
Observational Data ◽  
Radio Telescope ◽  
Dependence Analysis ◽  
Scientific Program ◽  
Radio Telescopes ◽  
Space Radio Telescope ◽  
Visibility Function ◽  
Maser Spot ◽  
Project Structure

Within the Radioastron interferometer scientific program, observational data of the H2O maser at a frequency of 22.2280 GHz in the NGC 2071 nebula were processed. A space radio telescope (SRT — 10 m) and three radio telescopes of the ground network: RT — 32 m (Medicina, Italy), RT — 32 m (Torun, Poland) and RT —64 m (Kalyazin, RF) took part in the observations. A map of the maser spot distribution has been obtained, where there are 13 spatial components with VLSR in the range 4.7 — 20.5 km/s. Correlation is observed on ground-space baselines for the component on VLSR = 14.3 km/s. Based on the visibility function dependence analysis from the baseline projection values, there was proposed a twocomponent model of this component spatial structure with the dimensions of the extended and compact constituent of 4 and 0.06 msec, i.e. 1.56 and 0.023 au, respectively.

Direction Dependence Analysis

2020 ◽  
pp. 9-46
Author(s):  
Wolfgang Wiedermann ◽  
Xintong Li ◽  
Alexander Eye
Keyword(s):  
Dependence Analysis ◽  
Direction Dependence

Conditional direction dependence analysis in linear models with SPSS macros and custom dialogue implementation

2018 ◽  
Author(s):  
◽  
Xintong Li
Keyword(s):  
Linear Models ◽  
Causal Effect ◽  
Dependence Analysis ◽  
University Of Missouri ◽  
Worked Example ◽  
Purification Technique ◽  
Conditional Effect ◽  
Illustrative Purpose ◽  
Direction Dependence ◽  
The University

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] In nonexperimental data, the causal ordering of variables can be examined with Direction dependence analysis (DDA), a statistical method that utilizes various asymmetry properties of the linear regression model to validate a postulated explanatory model against plausible causally reversed alternative models. However, standard DDA assumes that the observed causal effect is constant for all subjects and does not consider the conditional effect of a third variable on direction dependence, which may lead to biases and/or compromised power. The present work relaxes this assumption by proposing conditional direction dependence analysis (CDDA). CDDA examines the direction of effect when a moderator is present and extends standard DDA by combining the pick-a-point approach and variable purification technique, which enables researchers to examine the direction of effect at a certain moderator value. The results of two Monte-Carlo simulation studies are reported which evaluate the performance of CDDA. The first simulation study shows that the observed power of DDA tests vary across moderator values when a t hird variable moderators the main effect. The second study shows that, under certain conditions, CDDA is able to identify the true data-generating mechanism when a third variable determines the direction of causal flow. SPSS macros and auxiliary custom dialogues are provided for easy implementation of CDDA procedures, which is illustrated with a worked example. A real-world example is given for illustrative purpose.

Sign in / Sign up

Export Citation Format

Share Document