Large-Scale Survey Data Analysis with Penalized Regression: A Monte Carlo Simulation on Missing Categorical Predictors

Yoo, Jin Eun; Rho, Minjeong

doi:10.1080/00273171.2021.1891856

Cited by 7 publications

(19 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beyond producing interpretable models, the Enet and Mnet models of this study were comparable to RF models in terms of prediction. Likewise, multiple studies across diverse disciplines reported that linear models are comparable to RF (e.g., [53]- [55]) or even better than RF (e.g., [19], [20], [56]). These studies with ours have in common that the variables were pre-selected based on previous research.…”

Section: A Regularization and Learning Analyticsmentioning

confidence: 98%

“…Regularization produces biased estimates, and significance testing is performed on unbiased estimates. Special techniques such as post-selection inference (e.g., [43]) are required to perform statistical testing after regularization, but currently only available with LASSO [19], [20]. Instead of statistical testing, we iterated data splitting and prediction modeling, and obtained selection counts as the criterion for variable importance; variables selected more often bear more importance than variables selected less often [19], [20].…”

Section: ) Cross-validation (Cv)mentioning

confidence: 99%

“…Specifically, this study presented variables selected 1, 250, 500, 750 times or more, and all 1,000 times. Yoo and Rho [20] suggest that variables selected 75% or more with Enet and 50% or more with Mnet are worth investigation in a simulation study of social science large-scale data, but there is no such study in the context of learning analytics. All the programs were written in R 3.6.2.…”

Section: ) Cross-validation (Cv)mentioning

confidence: 99%

“…We propose regularization (penalized regression) among ML to analyze LMS log data. There has been little research employing regularization methods in LMS log data analysis, but regularization can contribute to the field in that it produces interpretable prediction models [18]- [20]. Based on linear regression, the regression coefficients of regularization can be interpreted similarly to those in traditional, non-penalized regression.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Online Students’ Learning Behaviors and Academic Success: An Analysis of LMS Log Data From Flipped Classrooms via Regularization

2022

Self Cite

View full text Add to dashboard Cite

The main purpose of this study was to demonstrate the uses of regularization, a machine learning technique, in exploring important predictors for online student success. We analyzed student and learning behavioral variables from undergraduate fully-online flipped classrooms. In particular, students' instructional video watching behaviors at an instructional unit level were extracted from LMS (learning management system) log data, and Enet (elastic net) and Mnet were employed among regularization. As results, regularization not only showed comparable prediction performance to random forest, a nonlinear method well-known for its prediction capabilities, but also produced interpretable prediction models as a linear method. Enet and Mnet selected 17 and 19 important predictors out of 159, respectively, and could identify potential low-performers as early as the first instructional week of the course. Important variables rarely recognized in previous studies included complete viewings of the first video before class and repeated complete viewings of challenging contents after in-class meetings. Unlike previous studies, aggregate measures of video lecture views were not important predictors. Variables less frequently studies in previous studies were the number of non-mandatory quiz-taking and mobile lecture watching frequencies. Variables in line with previous research were student attitudes towards the course, gender, grade level, and clicks on learning materials postings. Many students turned out not to watch lecture videos completely before class. Further research on regularization and exploration of these variables with other potentially important predictors can provide more insight into students' online learning from a comprehensive perspective.

show abstract

Section: A Regularization and Learning Analyticsmentioning

confidence: 98%

Section: ) Cross-validation (Cv)mentioning

confidence: 99%

Section: ) Cross-validation (Cv)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Online Students’ Learning Behaviors and Academic Success: An Analysis of LMS Log Data From Flipped Classrooms via Regularization

2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…To this end, based on measurements collected by the sensor network of a photovoltaic production plant, the paper proposes Monte Carlo (MC) simulation as the pre-processing stage to deal with outliers before applying PCA [4,5]. In this respect, the proposed approach is shown to be a valid alternative to relying on the classical Interquartile Range (IQR) method in order to omit outliers when applying PCA for anomaly detection purposes.…”

Section: Introductionmentioning

confidence: 99%

Anomaly Detection in Photovoltaic Production Factories via Monte Carlo Pre-Processed Principal Component Analysis

Arena

Corsini

Ferulano³

et al. 2021

Energies

View full text Add to dashboard Cite

This paper investigates a use case of robust anomaly detection applied to the scenario of a photovoltaic production factory—namely, Enel Green Power’s 3SUN solar cell production plant in Catania, Italy—by considering a Monte Carlo based pre-processing technique as a valid alternative to other typically used methods. In particular, the proposed method exhibits the following advantages: (i) Outlier replacement, by contrast with traditional methods which are limited to outlier detection only, and (ii) the preservation of temporal locality with respect to the training dataset. After pre-processing, the authors trained an anomaly detection model based on principal component analysis and defined a suitable key performance indicator for each sensor in the production line based on the model errors. In this way, by running the algorithm on unseen data streams, it is possible to isolate anomalous conditions by monitoring the above-mentioned indicators and virtually trigger an alarm when exceeding a reference threshold. The proposed approach was tested on both standard operating conditions and an anomalous scenario. With respect to the considered use case, it successfully anticipated a fault in the equipment with an advance of almost two weeks, but also demonstrated its robustness to false alarms during normal conditions.

show abstract

Investigating Variable Selection Techniques Under Missing Data: A Simulation Study

Bain,

Shi

2024

Springer Proceedings in Mathematics &Amp; Statistics

View full text Add to dashboard Cite

Large-Scale Survey Data Analysis with Penalized Regression: A Monte Carlo Simulation on Missing Categorical Predictors

Cited by 7 publications

References 71 publications

Online Students’ Learning Behaviors and Academic Success: An Analysis of LMS Log Data From Flipped Classrooms via Regularization

Online Students’ Learning Behaviors and Academic Success: An Analysis of LMS Log Data From Flipped Classrooms via Regularization

Anomaly Detection in Photovoltaic Production Factories via Monte Carlo Pre-Processed Principal Component Analysis

Investigating Variable Selection Techniques Under Missing Data: A Simulation Study

Contact Info

Product

Resources

About