A Permutation-Based Changepoint Technique for Monitoring Effect Sizes

2021 ◽  
pp. 1-12
Author(s):  
Daniel Kent ◽  
James D. Wilson ◽  
Skyler J. Cranmer
Keyword(s):  
Null Distribution ◽  
Permutation Testing ◽  
Statistical Process ◽  
Cross Sectional ◽  
Stochastic Fluctuations ◽  
Data Generating Process ◽  
Changepoint Analysis ◽  
Dynamic Relationships ◽  
Statistical Process Monitoring ◽  
Time Invariant

Abstract Across the social sciences, scholars regularly pool effects over substantial periods of time, a practice that produces faulty inferences if the underlying data generating process is dynamic. To help researchers better perform principled analyses of time-varying processes, we develop a two-stage procedure based upon techniques for permutation testing and statistical process monitoring. Given time series cross-sectional data, we break the role of time through permutation inference and produce a null distribution that reflects a time-invariant data generating process. The null distribution then serves as a stable reference point, enabling the detection of effect changepoints. In Monte Carlo simulations, our randomization technique outperforms alternatives for changepoint analysis. A particular benefit of our method is that, by establishing the bounds for time-invariant effects before interacting with actual estimates, it is able to differentiate stochastic fluctuations from genuine changes. We demonstrate the method’s utility by applying it to a popular study on the relationship between alliances and the initiation of militarized interstate disputes. The example illustrates how the technique can help researchers make inferences about where changes occur in dynamic relationships and ask important questions about such changes.

scholarly journals Proposing a machine-learning based method to predict stillbirth before and during delivery and ranking the features: nationwide retrospective cross-sectional study

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Toktam Khatibi ◽  
Elham Hanifi ◽  
Mohammad Mehdi Sepehri ◽  
Leila Allahqoli
Keyword(s):  
Machine Learning ◽  
External Validation ◽  
Fetal Loss ◽  
Null Distribution ◽  
Training Dataset ◽  
Gradient Boosting ◽  
Support Vector ◽  
Cross Sectional ◽  
Boosting Method ◽  
Demographic Features

Abstract Background Stillbirth is defined as fetal loss in pregnancy beyond 28 weeks by WHO. In this study, a machine-learning based method is proposed to predict stillbirth from livebirth and discriminate stillbirth before and during delivery and rank the features. Method A two-step stack ensemble classifier is proposed for classifying the instances into stillbirth and livebirth at the first step and then, classifying stillbirth before delivery from stillbirth during the labor at the second step. The proposed SE has two consecutive layers including the same classifiers. The base classifiers in each layer are decision tree, Gradient boosting classifier, logistics regression, random forest and support vector machines which are trained independently and aggregated based on Vote boosting method. Moreover, a new feature ranking method is proposed in this study based on mean decrease accuracy, Gini Index and model coefficients to find high-ranked features. Results IMAN registry dataset is used in this study considering all births at or beyond 28th gestational week from 2016/04/01 to 2017/01/01 including 1,415,623 live birth and 5502 stillbirth cases. A combination of maternal demographic features, clinical history, fetal properties, delivery descriptors, environmental features, healthcare service provider descriptors and socio-demographic features are considered. The experimental results show that our proposed SE outperforms the compared classifiers with the average accuracy of 90%, sensitivity of 91%, specificity of 88%. The discrimination of the proposed SE is assessed and the average AUC of ±95%, CI of 90.51% ±1.08 and 90% ±1.12 is obtained on training dataset for model development and test dataset for external validation, respectively. The proposed SE is calibrated using isotopic nonparametric calibration method with the score of 0.07. The process is repeated 10,000 times and AUC of SE classifiers using random different training datasets as null distribution. The obtained p-value to assess the specificity of the proposed SE is 0.0126 which shows the significance of the proposed SE. Conclusions Gestational age and fetal height are two most important features for discriminating livebirth from stillbirth. Moreover, hospital, province, delivery main cause, perinatal abnormality, miscarriage number and maternal age are the most important features for classifying stillbirth before and during delivery.

Statistical process monitoring for e-waste based on beta regression and particle swarm optimization

2022 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Angelo Marcio Oliveira Sant’Anna
Keyword(s):  
Particle Swarm Optimization ◽  
Regression Model ◽  
Process Monitoring ◽  
Manufacturing Process ◽  
Particle Swarm ◽  
Injection Rate ◽  
Statistical Process ◽  
Swarm Optimization ◽  
Content Type ◽  
Statistical Process Monitoring

PurposeE-waste management can reduce relevant impact of the business activity without affecting reliability, quality or performance. Statistical process monitoring is an effective way for managing reliability and quality to devices in manufacturing processes. This paper proposes an approach for monitoring the proportion of e-waste devices based on Beta regression model and particle swarm optimization. A statistical process monitoring scheme integrating residual useful life techniques for efficient monitoring of e-waste components or equipment was developed.Design/methodology/approachAn approach integrating regression method and particle swarm optimization algorithm was developed for increasing the accuracy of regression model estimates. The control chart tools were used for monitoring the proportion of e-waste devices from fault detection of electronic devices in manufacturing process.FindingsThe results showed that the proposed statistical process monitoring was an excellent reliability and quality scheme for monitoring the proportion of e-waste devices in toner manufacturing process. The optimized regression model estimates showed a significant influence of the process variables for both individually injection rate and toner treads and the interactions between injection rate, toner treads, viscosity and density.Originality/valueThis research is different from others by providing an approach for modeling and monitoring the proportion of e-waste devices. Statistical process monitoring can be used to monitor waste product in manufacturing. Besides, the key contribution in this study is to develop different models for fault detection and identify any change point in the manufacturing process. The optimized model used can be replicated to other Electronic Industry and allows support of a satisfactory e-waste management.

scholarly journals ARE ISLAMIC STOCKS LESS EXPOSED TO SENTIMENT-BASED MISPRICING THAN NON-ISLAMIC ONES? EVIDENCE FROM THE INDONESIAN STOCK EXCHANGE

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Rizqi Umar Al Hashfi ◽  
Ahmad Maulin Naufa ◽  
U’um Munawaroh
Keyword(s):  
Capital Market ◽  
Stock Exchange ◽  
Investor Sentiment ◽  
Principal Component ◽  
Equation Model ◽  
Cross Sectional ◽  
Stock Mispricing ◽  
Time Invariant ◽  
Islamic Values

The aim of this research is to verify the role of Islamic value in stock mispricing in the Indonesian capital market. Empirically, high investor sentiment can lead to mispricing on equity appraisal. When investors feel excessively optimistic about their valuation, equity will be overpriced, or vice versa. The presence of Islamic values, such as the prohibition of interest, speculative and uncertain transactions, and excessive leverage, arguably reduce sentiment-based mispricing. Daily and cross-sectional market data were employed. In addition, principal component analysis was conducted to construct a firm-specific investor sentiment variable. With regard to the method, the Hausman-Taylor (H-T) approach was used to deal with heterogeneity, endogeneity, and the time-invariant variable in Fama-MacBeth regression. The results show that our baseline analysis confirms the mispricing of overall stocks. However, Islamic stocks are less exposed to sentiment-based mispricing than their non-Islamic counterparts. The results are consistent with our robustness test, in which we estimate the equation model across industry and portfolio. Finally, our findings imply various insights for both investors and policymakers.

Sign in / Sign up

Export Citation Format

Share Document