Structural Equation Model Averaging: Methodology and Application

Seng, Loraine; Li, Jialiang

doi:10.1080/07350015.2020.1870479

Cited by 13 publications

(10 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likewise, the corresponding least square estimate of the Kn × 1 weight vector replacing Equation ( 9) is Wn = e Y ⊥ . Since the post-selection procedure of estimating the weight vector of the selected set of submodels is the same as the pre-selection procedure given in Seng and Li, 49 the asymptotic normality and consistency of the effect estimator follows. Adjusting to the notation used in this article, for using the Kn submodels' predictions in the final estimator, as sample size n goes to infinity,…”

Section: Ijmentioning

confidence: 99%

“…This article seeks to extend the model average method 48,49 to high-dimensional instruments data when both the number of models and the individual model size could diverge with the sample size. One may treat each submodel's prediction as a composite IV.…”

Section: Introductionmentioning

confidence: 99%

“…In Seng and Li, 49 K n is assumed to be fixed and does not diverge with n. In such case, the optimal weight W n0 = (w 01 , … , w 0K n ) T for a selected set of models is defined as…”

mentioning

confidence: 99%

“…The consistency of this weight estimator (9) been established in Seng and Li. 49 In this work we do not require the weights to be bounded in [0, 1] and sum to one. Earlier authors 29,30,55 recommended such unconstrained estimation to achieve better prediction results.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Instrumental variable model average with applications in Mendelian randomization

Seng

Liu

Wang

et al. 2023

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

Mendelian randomization is a technique used to examine the causal effect of a modifiable exposure on a trait using an observational study by utilizing genetic variants. The use of many instruments can help to improve the estimation precision but may suffer bias when the instruments are weakly associated with the exposure. To overcome the difficulty of high‐dimensionality, we propose a model average estimator which involves using different subsets of instruments (single nucleotide polymorphisms, SNPs) to predict the exposure in the first stage, followed by weighting the submodels' predictions using penalization by common penalty functions such as least absolute shrinkage and selection operator (LASSO), smoothly clipped absolute deviation (SCAD) and minimax concave penalty (MCP). The model averaged predictions are then used as a genetically predicted exposure to obtain the estimation of the causal effect on the response in the second stage. The novelty of our model average estimator also lies in that it allows the number of submodels and the submodels' sizes to grow with the sample size. The practical performance of the estimator is examined in a series of numerical studies. We apply the proposed method on a real genetic dataset investigating the relationship between stature and blood pressure.

show abstract

Section: Ijmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…In Seng and Li, 49 K n is assumed to be fixed and does not diverge with n. In such case, the optimal weight W n0 = (w 01 , … , w 0K n ) T for a selected set of models is defined as…”

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Instrumental variable model average with applications in Mendelian randomization

Seng

Liu

Wang

et al. 2023

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

show abstract

“…(4) The relationship between the land expropriation procedure and farmers' satisfaction The latent variable land expropriation procedure was positively related to the observation variables, including land expropriation announcement, formulation of land expropriation plan, public opinion feedback, and contradiction mediation. Among them, the coefficients of formulation of the land expropriation plan (0.73) and contradiction mediation (0.73) were the highest, indicating that in the process of land expropriation, we should focus on the land expropriation scheme and the emergency plan to solve the conflicts around land expropriation [62][63][64][65].…”

Section: Relationship Between Latent Variables and Farmers' Satisfactionmentioning

confidence: 99%

Farmers’ Satisfaction with Land Expropriation System Reform: A Case Study in China

et al. 2021

Land

View full text Add to dashboard Cite

Farmers’ satisfaction with reform of the land expropriation system has not been fully examined, so it is difficult to comprehensively and successfully judge the effectiveness of the reforms. Traditional statistical methods cannot accurately explain the relationship between the variables. In order to fully understand the implementation, progress, and applicability of land expropriation system reform, this paper analyzes the factors influencing farmers’ satisfaction, presents the shortcomings of land expropriation system reform, and puts forward improvement suggestions. Taking the land expropriation system reform pilot in Dingzhou city as an example, this paper investigates the satisfaction of the farmers who have had their land expropriated by establishing a structural equation model (SEM) to obtain feedback on the implementation effect of the pilot work. The results show that the factors affecting farmers’ satisfaction with the reform of land expropriation systems can be summarized into four variables: land expropriation compensation standard, land income distribution, land expropriation security mode, and land expropriation procedure. The parameter estimation between these four potential variables and their corresponding observation variables shows that, in the process of land expropriation, the comparison of land expropriation compensation standards and observation variables with market entry projects is an important factor affecting the satisfaction of farmers with regard to land expropriation. The income of municipal and county governments and village collective incomes have a great impact on farmers, indicating that the satisfaction of farmers is not only related to the absolute level of compensation and income, but is also affected by the relative level. In addition, providing farmers with a variety of reasonable security methods is an important element to enable the smooth progress of land expropriation. Furthermore, attention should also be paid to the formulation of land expropriation schemes and emergency plans to solve land expropriation conflicts.

show abstract

Segment regression model average with multiple threshold variables and multiple structural breaks

Liu

2023

Can J Statistics

Self Cite

View full text Add to dashboard Cite

We propose a new model averaging approach to investigate segment regression models with multiple threshold variables and multiple structural breaks. We first fit a series of models, each with a single threshold variable and multiple breaks over its domain, using a two‐stage change point detection method. Then these models are combined together to produce a weighted ensemble through a frequentist model averaging approach. Consequently, our segment regression model averaging (SRMA) method may help identify complicated subgroups in a heterogeneous study population. A crucial step is to determine the optimal weights in the model averaging, and we follow the familiar non‐concave penalty estimation approach. We provide theoretical support for SRMA by establishing the consistency of individual fitted models and estimated weights. Numerical studies are carried out to assess the performance in low‐ and high‐dimensional settings, and comparisons are made between our proposed method and a wide range of existing alternative subgroup estimation methods. Two real economic data examples are analyzed to illustrate our methodology.

show abstract

Structural Equation Model Averaging: Methodology and Application

Cited by 13 publications

References 49 publications

Instrumental variable model average with applications in Mendelian randomization

Instrumental variable model average with applications in Mendelian randomization

Farmers’ Satisfaction with Land Expropriation System Reform: A Case Study in China

Segment regression model average with multiple threshold variables and multiple structural breaks

Contact Info

Product

Resources

About