A convex formulation for path analysis in structural equation modeling

Pruttiakaravanich, Anupon; Songsiri, Jitkomut

doi:10.1109/sice.2016.7749179

Cited by 2 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The important theoretical results that α cannot be arbitrarily large as the solution X 1 in (6) becomes an impracticable approximate of Σ −1 in the original problem of (3) [PS16]. For the completeness of this paper, we state the proposition in [PS16] and provide the proof here. Let us start with the zero gradient condition of Karush-Kuhn-Tucher (KKT): X 1 = (S − Z) −1 , when Z = 0 is an optimal dual.…”

Section: Trivial Dual Solutionsmentioning

confidence: 96%

“…Previous work [PS16] applied a convex relaxation to the quadratic equality constraint of (3) and proposed the convex confirmatory SEM formulation minimize − log det X 1 + tr(SX 1 ) subject to X 0, 0 X 4 αI, P (X 2 ) = I,…”

Section: Confirmatory Semmentioning

confidence: 99%

“…Previous analysis [PS16] showed that when an optimal X has rank n, then In Section 3.1 it was shown that the minimum eigenvalue of S always lies on the left of the harmonic mean of the eigenvalues of S. If we choose α = λ min (S), then it is often the case that rank(Z) = n. This advises us to include a constraint X 4 λ min (S)I S into the estimation problem. Since X 4 = Ψ, we can justify that the covariance error is controlled to be less than the covariance of the variables.…”

Section: Low Rank Solutionmentioning

confidence: 99%

See 2 more Smart Citations

Convex formulation for regularized estimation of structural equation models

Pruttiakaravanich

Songsiri

2020

Signal Processing

Self Cite

View full text Add to dashboard Cite

Path analysis is a model class of structural equation modeling (SEM), which it describes causal relations among measured variables in the form of a multiple linear regression. This paper presents two estimation formulations, one each for confirmatory and exploratory SEM, where a zero pattern of the estimated path coefficient matrix can explain a causality structure of the variables. The original nonlinear equality constraints of the model parameters were relaxed to an inequality, allowing the transformation of the original problem into a convex framework. A regularized estimation formulation was then proposed for exploratory SEM using an l1-type penalty of the path coefficient matrix. Under a condition on problem parameters, our optimal solution is low rank and provides a useful solution to the original problem. Proximal algorithms were applied to solve our convex programs in a large-scale setting. The performance of this approach was demonstrated in both simulated and real data sets, and in comparison with an existing method. When applied to two real application results (learning causality among climate variables in Thailand and examining connectivity differences in autism patients using fMRI time series from ABIDE data sets) the findings could explain known relationships among environmental variables and discern known and new brain connectivity differences, respectively.

show abstract

Section: Trivial Dual Solutionsmentioning

confidence: 96%

Section: Confirmatory Semmentioning

confidence: 99%

Section: Low Rank Solutionmentioning

confidence: 99%

See 1 more Smart Citation

Convex formulation for regularized estimation of structural equation models

Pruttiakaravanich

Songsiri

2020

Signal Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this regard, path analysis can be employed. Path analysis is Structural Equation Modeling (SEM), the model formed describes the causal relationship between measured variables in the form of multiple linear regression [1]. Measurements can be made based on the consumer's assessment of the performance of a service [2], [3].…”

Section: Introductionmentioning

confidence: 99%

“…Path analysis is Structural Equation Modeling (SEM), the model formed describes the causal relationship between measured variables in the form of multiple linear regression [1]. Measurements can be made based on the consumer's assessment of the performance of a service [2], [3]. There are seven dimensions of service quality of academic namely, administrative services, tangibles, academic programs, academic staff, teaching delivery, certainty, and empathy [4].…”

Section: Introductionmentioning

confidence: 99%

Student Service Performance Analysis by Using Path Analysis Model

2019

IJRTE

View full text Add to dashboard Cite

Academic services are probably the most important aspect to achieve excellence in a university. Every university applies strategies to improve their service quality for example an integrated academic information system that has been run by Universitas Muslim Indonesia (UMI). Due to the students as the object of services, it is important to evaluate their level of satisfaction regarding to the academic service performance that they have experienced. This research aimed to analyze the students’ satisfaction on a WEB based academic information system in UMI using path analysis. The research phases started from determining hypothesis, collecting questionnaire data, identifying variables, analyzing data using path analysis, and analyzing model. The variables tested were students’ registration, course schedule, course grades and assistantships. In this research, a model was created to measure the level of users’ satisfaction by information system based services. The results of this research is the level of satisfaction in student registration services that higher than the course schedule, course grades, and academic assistantships and a gain of confidence to the student that is equal to 17%. Academic information system (SIMAK) on academic service system of UMI can be analyzed by using path analysis and obtained student satisfaction result 69%..

show abstract

A convex formulation for path analysis in structural equation modeling

Cited by 2 publications

References 10 publications

Convex formulation for regularized estimation of structural equation models

Convex formulation for regularized estimation of structural equation models

Student Service Performance Analysis by Using Path Analysis Model

Contact Info

Product

Resources

About