Open-source Statistical Software Research Articles

Advanced machine learning for missing petrophysical property imputation applied to improve the characterization of carbonate reservoirs

Missing data is a common cause of uncertainty in reservoir characterization, especially in core analysis of porosity and permeability. This is due to the high cost and time required to gather core samples while drilling through the entire reservoir thickness. Therefore, the ability to collect core samples and directly assess petrophysical parameters is restricted by the small proportion of the reservoir thickness that can be measured by the retrieved cores. As a result, the absence of data has the potential to lead to inconsistent and imprecise classification of reservoirs and geomodeling, thereby increasing reservoir uncertainty. Therefore, imputation algorithms are needed to estimate petrophysical properties for incomplete intervals. This study comprehensively compares seven imputation techniques for predicting missing horizontal and vertical core permeability and core porosity data in a well drilled through a carbonate reservoir in a southern Iraqi gas field. The performance of each method was assessed using relative bias (RB) and reasonable descriptive statistical distributions (histogram, log view, and box plots of data ratios to compare the imputed and original data), and robustness to outliers (RtO). The results revealed that the random imputation of missing data (RIMD) technique is the most effective algorithm in addressing the 30.5% missing data (vertical permeability), resulting in the attainment of the RB value closest to zero (0.04076191) compared to other algorithms. Principal component analysis (PCA) and random forest (RF) yielded the most favorable results in terms of handling missing values and outliers for the 15% missing ratio dataset (core porosity and horizontal permeability). The advantages of these approaches are related to their capacity to independently and accurately impute the missing data without requiring extra information, such as well logging records. Additionally, the multiple imputation and machine learning algorithms fill missing values with adequate numbers based on dataset variable distributions and correlations. Multiple plausible values quantify the uncertainty in estimating missing data, avoiding false precision (as with single imputation). Also, the novel workflow involving multiple imputation methods accurately take into account the correlations and variances between the variables. The novel workflow developed for missing data imputation has the potential to be applied to other clastic and carbonate reservoirs in oil and gas fields. The entire data imputation workflow was conducted, validated, visualized by codes in R, a robust open-source statistical computing software.

IJCM_329A: Prevalence of Impostor Phenomenon and its association with Depression, Anxiety and Stress in undergraduate medical students

Background: Clance and Imes coined the term “impostor phenomenon,” or IP, which causes a person to choose to elevate the achievements of others and belittle their own successes. This results in constant self-underappreciation and anxiety about a task. The field of medical science is a highly pressure-intensive and competitive field. In the medical field, where patients’ lives are involved daily, a condition like this which is going undiagnosed, will affect the duties provided by the medical practitioner in the long run. They will doubt every decision they make, which can seriously harm the patient. All these reasons show the necessity of discovering new and more advanced techniques to identify the disorder. As a result, this condition develops into more severe forms. It results in more serious psychiatric conditions that can be prevented if IP is identified correctly. Objective: To identify the prevalence of Imposter Phenomenon and to find its associations with depression, anxiety, and stress in medical students Methodology: It was a cross sectional study done among undergraduate students of a private medical college in Manipal. About 377 undergraduates participated in the study. Data was collected using the Clance IP scale and DASS-21 questionnaire after taking relevant permissions. The data was collected in an excel sheet from the responses derived from the Microsoft form. Data was analyzed using the open-source statistical software Jamovi 2.0. Results: It was observed that more than half of the students were female. More than two fifths of the students reported having frequent IP experiences, and about 15% of them reported intense IP experiences. Positive correlation was observed between IP and Stress, Anxiety and Depression. Conclusion: Impostor Phenomenon should be diagnosed and identified early in medical undergraduates to reduce mental health issues in the long run so they may be able to provide quality medical care.

An investigation of endometrial vascularity in normally menstruating women and those in other physiological states using 3D ultrasound imaging.

This study aimed to determine the normal vasculature indices of the endometrium and to correlate them with those in various physiological states. Women undergoing ultrasound at the Feto-Maternal Center, Qatar in 2020-2021 as part of their gynecologic evaluation were enrolled into the study. They were divided into those with normal menses and no additional pathology, those following spontaneous miscarriage, postpartum and menopausal. Three-dimensional (3D) evaluation of the endometrial vasculature was done and the parameters quantified included vascularization index (VI), flow index (FI), vascularization flow index (VFI), endometrial thickness, endometrial volume and uterine volume. JASP, an open-source statistical analysis software, was used for analysis and an independent t-test to compare the vascularity indices. A multivariate regression analysis was also done to look at the factors affecting the endometrial vascular indices within the luteal phase. A total of 461 women were studied: 122 in the follicular phase, 199 in the luteal phase, 90 after a spontaneous miscarriage, 29 postpartum, and 16 menopausal. The vascularity indices were highest after miscarriage and lowest postnatally. There were no significant effects of age, gravida, para, or abortions on VI and VFI. However, there was a significant positive effect of age on FI (P = 0.019) There was a significant increase in endometrial volume and thickness in the luteal phase as compared to follicular phase (P < 0.01), but there was no difference in the vascularity indices. The uterine and endometrial volume in the postnatal group were nearly double that of the luteal group (P value <0.01 and 0.014, respectively). There was a significant decrease in flow index in the postnatal group compared to the luteal group (P < 0.01), suggesting low flow intensity in the postnatal group. Endometrial vascular indices measured using 3D Doppler can be used to determine normal vascular indices and vary with physiological states such as after miscarriages, postnatally and in the menopausal states.

Overseeing Infill

Problem, research strategy, and findings Several U.S. states with high housing costs have recently adopted laws intended to promote infill development. These new laws expand state agencies’ supervisory responsibilities to ensure that local governments comply with state mandates. Effective administration of these laws will require state agencies to accurately estimate the amount of new housing that might be created and to target review to the jurisdictions that are failing to meet the relevant requirements. Here we present quantitative tools both for prioritizing review of local plans and zoning ordinances and for estimating future housing development. We applied the tools to the implementation of California laws requiring local governments to amend their zoning ordinances to allow accessory dwelling units on parcels zoned for detached single-family housing development. We provide computer code, written in the open-source statistical computing language R, that implements these tools. Although we present off-the-shelf tools, our proposed tools should supplement other regulatory techniques rather than serving as a substitute. Takeaway for practice Requirements for local governments to allow infill development should be accompanied by mandates for data collection. With good data, state agencies can use open-source statistical software to create quantitative measures that can help estimate future housing production and set priorities for reviewing local plans and zoning ordinances.

Quantitative Political Science Education in the Past and Future

There has been a massive shift in teaching quantitative political research since the Journal of Political Science Education was launched in 2004. Smartphones were an anomaly, and it was uncommon to have laptops in the classroom. Statistical calculations were sometimes done by “statisticians”, i.e., professional staff who did calculations for faculty members. Today, it is rare to see students without electronics. Through that transition we experienced ubiquitous Wi-Fi and smartphones, statistical computing on personal computers, the end of the academic staff statistician, an explosion in open-source statistical software and tutorials, and an unexpected mass transition to online learning during COVID. We experienced a similar revolution in teaching statistics. Increases in computational power and data availability make quantitative and qualitative research different than 20 years ago. Computation is rarely a limiting factor, and we find ourselves spending more time on statistical assumptions, correct methods, data integrity, and replicability. We are now entering an era of assistive technology and will need to transition to teaching students how to use artificial intelligence tools to assist them with quantitative work. In this article, we consider these changes and what they mean for teaching political science in the next 20 years.

Explanatory item response models for instrument validation: A tutorial based on an elicited imitation test

This tutorial introduces explanatory item response models for the purpose of validating instruments used in applied linguistics research. These models can be used to investigate how theoretically important, construct-relevant characteristics of items influence the difficulty associated with successful responses, which constitutes valuable validity evidence. We focus on two item-explanatory models, the linear logistic test model for dichotomous responses and the linear rating scale model for ordinal responses, in the context of elicited imitation tests commonly used to measure oral proficiency in L2 research. Examples are provided using open data and the open-source statistical software R.

A Tutorial on Causal Inference in Longitudinal Data With Time-Varying Confounding Using G-Estimation

In psychological research, longitudinal study designs are often used to examine the effects of a naturally observed predictor (i.e., treatment) on an outcome over time. But causal inference of longitudinal data in the presence of time-varying confounding is notoriously challenging. In this tutorial, we introduce g-estimation, a well-established estimation strategy from the causal inference literature. G-estimation is a powerful analytic tool designed to handle time-varying confounding variables affected by treatment. We offer step-by-step guidance on implementing the g-estimation method using standard parametric regression functions familiar to psychological researchers and commonly available in statistical software. To facilitate hands-on usage, we provide software code at each step using the open-source statistical software R. All the R code presented in this tutorial are publicly available online.

Rigour and reproducibility in perinatal and paediatric epidemiologic research using big data.

Rigour and reproducibility in perinatal and paediatric epidemiologic research using big data.

Teacher Adaptability to Change in Special Education: Creativity and Commitment to the Institution

The transition to inclusive education implies a global transformation of the school. Mexican teachers face the implementation of the paradigm of inclusion in special education. We set out to define new skills needed in special education teachers' profiles in the inclusion framework. In the survey presented to 39 Special Education graduates in Puebla who responded to our questionnaire, we related the variables of demand in their work to adapt to technological and organizational changes, commitment and satisfaction with the institution, and teaching competence in creative innovation. We processed the data with the open-source statistical software PSPP to test our data's reliability and relate the variables employing Pearson's correlation coefficient. In order to meet the job requirements of adapting to technological and organizational changes, special education tea-chers must enhance their competence of creative innovation (Pearson's R= .77); and feel commit-ment and satisfaction with the institution they work in (Pearson's R= .73). Such commitment to and satisfaction with the institution require collaborative work with the rest of the teaching staff and the institution's authorities. Conditions for teachers' identification with and commitment to the institution build up through a dialectic relationship between the institutions and employees.

Bayesian tests of two proportions: A tutorial with R and JASP

The need for a comparison between two proportions (sometimes called an A/B test) often arises in business, psychology, and the analysis of clinical trial data. Here we discuss two Bayesian A/B tests that allow users to monitor the uncertainty about a difference in two proportions as data accumulate over time. We emphasize the advantage of assigning a dependent prior distribution to the proportions (i.e., assigning a prior to the log odds ratio). This dependent-prior approach has been implemented in the open-source statistical software programs R and JASP. Several examples demonstrate how JASP can be used to apply this Bayesian test and interpret the results.

MRI adipose tissue segmentation and quantification in R (RAdipoSeg)

BackgroundExcess adipose tissue is associated with increased cardiovascular and metabolic risk, but the volume of visceral and subcutaneous adipose tissue poses different metabolic risks. MRI with fat suppression can be used to accurately quantify adipose depots. We have developed a new semi-automatic method, RAdipoSeg, for MRI adipose tissue segmentation and quantification in the free and open source statistical software R.MethodsMRI images were obtained from wild-type mice on high- or low-fat diet, and from 20 human subjects without clinical signs of metabolic dysfunction. For each mouse and human subject, respectively, 10 images were segmented with RAdipoSeg and with the commercially available software SliceOmatic. Jaccard difference, relative volume difference and Spearman’s rank correlation coefficients were calculated for each group. Agreement between the two methods were analysed with Bland–Altman plots.ResultsRAdipoSeg performed similarly to the commercial software. The mean Jaccard differences were 10–29% and the relative volume differences were below ( ±) 20%. Spearman’s rank correlation coefficient gave p-values below 0.05 for both mouse and human images. The Bland–Altman plots indicated some systematic and proporitional bias, which can be countered by the flexible nature of the method.ConclusionRAdipoSeg is a reliable and low cost method for fat segmentation in studies of mice and humans.

77 Hands-on: Making Sense of big Data, Machine Learning, and Modeling

Abstract Advancements in precision livestock technology have resulted in an unprecedented amount of data being collected on individual animals. Often processing of these datasets can be time consuming, tedious, and prone to human-error if processed with conventional software. Open-source statistical software (e.g., R of Python) can provide users with tools to automate many data processing steps for compiling and aggregating data. However, the steps from data collection to processing and training machine learning (ML) models can be time intensive for those new to statistical programming, with few examples pertaining to livestock. The objectives of this hands-on training are: 1) introduce workshop participants to methods for streamlining data processing tasks in Python and R, 2) demonstrate and provide examples of compiling large accelerometer datasets for determining daily livestock behavior; 3) introduce a suite of classification algorithms and validation testing approaches for classifying accelerometer training datasets, and 4) utilize model predictions to estimate and analyze daily behavior for beef cattle. Real life example datasets and code will be provided to workshop attendees to demonstrate how to take raw accelerometer datasets through a finished machine learning analysis. An example of estimating daily energy expenditure for individual animals using behavior data will be provided to highlight linkages to potential rangeland nutrition modeling applications. To obtain maximum benefit from this workshop, participants should bring a portable laptop computer to the workshop and will be encouraged to load software and preview content from a shared cloud directory prior to this training.

ShinyRBase: Near real-time energy saving models using reactive programming

To document energy savings from retrofitting a building, a reliable baseline model is needed. The development and implementation of the baseline model is an important step in the measurement and verification (M&V) process. Usually, an energy analyst enters the stage, collects data, do the estimation and delivers the baseline model. The modeling work of the energy analyst is done on either a proprietary or open-source statistical software, often using a coding script. If stakeholders want an updated report on energy savings, the analyst must re-do the whole process, for example on a monthly basis. This workflow is based on an imperative programming paradigm. The analyst holds on to the code that performs the analysis and re-run the code when agreed upon. The consequence of this workflow is that stakeholders are dependent on the energy analyst and that updated energy savings results must be planned and scheduled. However, emerging M&V 2.0 technologies enables automation of the energy saving reports. This paper demonstrates how energy savings from retrofitting’s in the Norwegian food retail sector is continuously monitored and documented in a web application. The application is built using open-source tools where the baseline model is delivered through a reactive programming framework. As an energy savings baseline model, the Tao Vanilla benchmarking model (TVB) was set into production in the web application. The TVB is a linear regression model with well specified features, easy to interpret and has a history of excellent prediction performance. The proposed web application framework allows for a fast development cycle without any need-to-know web programming languages like HTML, CSS or JavaScript. The reactive framework delivers several advantages. First, the stakeholders will always have a current and real-time report on the savings. Second, complex methodologies are dynamically used by the end-user. Third, increased involvement by stakeholders and interaction with the analyst related to the methods used in the energy savings analysis leads to collaborative benefits such as faster disseminating of knowledge. These synergy effect leads to a better technical understanding from the end user perspective and enhanced practical understanding for the analyst. Finally, the paper presents an integrated look at the energy kWh savings versus the cost of the retrofitting’s.

High Resolution Treatment Effects Estimation: Uncovering Effect Heterogeneities with the Modified Causal Forest.

There is great demand for inferring causal effect heterogeneity and for open-source statistical software, which is readily available for practitioners. The mcf package is an open-source Python package that implements Modified Causal Forest (mcf), a causal machine learner. We replicate three well-known studies in the fields of epidemiology, medicine, and labor economics to demonstrate that our mcf package produces aggregate treatment effects, which align with previous results, and in addition, provides novel insights on causal effect heterogeneity. For all resolutions of treatment effects estimation, which can be identified, the mcf package provides inference. We conclude that the mcf constitutes a practical and extensive tool for a modern causal heterogeneous effects analysis.

A tutorial on Bayesian single-test reliability analysis with JASP

The current practice of reliability analysis is both uniform and troublesome: most reports consider only Cronbach’s α, and almost all reports focus exclusively on a point estimate, disregarding the impact of sampling error. In an attempt to improve the status quo we have implemented Bayesian estimation routines for five popular single-test reliability coefficients in the open-source statistical software program JASP. Using JASP, researchers can easily obtain Bayesian credible intervals to indicate a range of plausible values and thereby quantify the precision of the point estimate. In addition, researchers may use the posterior distribution of the reliability coefficients to address practically relevant questions such as “What is the probability that the reliability of my test is larger than a threshold value of .80?”. In this tutorial article, we outline how to conduct a Bayesian reliability analysis in JASP and correctly interpret the results. By making available a computationally complex procedure in an easy-to-use software package, we hope to motivate researchers to include uncertainty estimates whenever reporting the results of a single-test reliability analysis.

데이터마이닝 기법을 활용한 관광목적지 이미지와 온라인 관광정보의 지각된 유용성과의 관계분석

This study seeks to analyze the relationship of tourism destination images and perceived usefulness of potential tourists in English reviews published on TripAdvisor, a post-travel reviews website belonging to online tourism information. The research method utilizes R programs, an open-source statistical software, to apply text mining techniques to quantify unstructured text data. First, a total of 25,016 reviews from 20 Seoul tourism destinations of TripAdvisor were collected with data mining techniques to extract the words, and a simple regression analysis was conducted with each extracted word. To test the hypothesis, a negative binary regression was conducted. The results of this study are as follows. First, it was confirmed that the tourism destination images affect perceived usefulness. Second, we were able to see the relationship among the tourism destination images. Finally, based on the results of these empirical analyses, we present academic and practical implications.

Using R in Regulatory Toxicology.

Statistical analyses are an essential part of regulatory toxicological evaluations. While projects would be ideally monitored by both toxicologists and statisticians, this is often not possible in practice. Hence, toxicologists should be trained in some common statistical approaches but also need a tool for statistical evaluations. Due to transparency needed in regulatory processes and standard tests that can be evaluated with template approaches, the freely available open-source statistical software R may be suitable. R is a well-established software in the statistical community. The principal input method is via software code, which is both benefit and weakness of the tool. It is increasingly used by regulating authorities globally and can be easily extended by software packages, e.g., for new statistical functions and features. This manuscript outlines how R can be used in regulatory toxicology, allowing toxicologists to perform all regulatory required data evaluations in a single software solution. Practical applications are shown in case studies on simulated and experimental data. The examples cover a) Dunnett testing of treatment groups against a common control and in relation to a biological relevance threshold, assessing the test's assumptions and plotting the results; b) dose-response analysis and benchmark dose derivation for chronic kidney inflammation as a function of Pyridine; and c) graphical/exploratory data analysis of previously published developmental neurotoxicity data for Chlorpyrifos.

Abstract 14187: Differential Neutralization of Unfractionated and Low Molecular Weight Heparin by Andexanet Alfa

Introduction: Andexanet Alfa is an antidote for the neutralization of the bleeding effects of Direct Oral Anticoagulant (Direct Xa) agents such as Rivaroxaban and Apixaban. It represents a molecularly modified factor Xa decoy protein with high specificity for factor Xa inhibitors. Unfractionated heparin (UFH) and low molecular weight heparins (LMWHs) exhibit both anti-Xa and anti-IIa activities. The purpose of this study is to investigate the relative neutralization of the anti-coagulant effects UFH and enoxaparin by andexanet alfa in whole blood assays such as activated clotting time (ACT) and thromelastography (TEG). Methods: The neutralization profiles of UFH and enoxaparin were studied by Andexanet at various concentrations. The final concentration of UFH used for activated clotting time (ACT) was 10 μg/ml, and for enoxaparin was 25 μg/ml. Andexanet was used in a concentration range of 12.5 ug/ml. For the thromboelastographic (TEG) analysis the concentration of all drugs were proportionately reduced. The results were analyzed using R open source statistical software. In order to compare the means of the groups, analysis of variance (ANOVA) tests were performed. If the ANOVA test yielded a significant result, a Tukey post-hoc analysis was performed. A pearson correlation was run for each parameter with respect to the dose of the drug. Results: Andexanet at 200 ug/ml moderately neutralized UFH (302 s vs 198). Andexanet minimally neutralized enoxaparin (200 s vs 190 s). There was no concentration dependent change in the neutralization profiles of UFH and enoxaparin at a concentration range of 12.5 ug/ml - 200 ug/ml. In the TEG assays, Andexanet alfa at 10 ug/ml partially neutralized the anticoagulant effects of UFH at 1 ug/ml as measured by various TEG parameters. At 20 ug/ml, stronger inhibition of all parameters were noted. Discussion: These studies suggest that Andexanet Alfa may be an effective neutralizing agent for UFH, however, it is ineffective in neutralizing LMWH at the concentrations studied. It is interesting to note that LMWHs have a higher anti-Xa to IIa ratio in comparison to heparin, however, their neutralization is lesser with Andexanet. These studies suggest that andexanet can be used in combination with protamine sulfate to neutralize UFH.

Using Rate Transient Analysis and Bayesian Algorithms for Reservoir Characterization in Unconventional Gas Wells during Linear Flow

Summary The main objective of this work is to improve robust, repeatable interpretation of reservoir characteristics using rate transient analysis (RTA). This is to generate probabilistic credible intervals for key reservoir and completion variables. This resulting data-driven algorithm was applied to production data from both synthetic and actual case histories. Synthetic production data from a multistage, hydraulically fractured horizontal completion in a reservoir modeled after the Marcellus Shale reservoir were generated using a reservoir model. The synthetic production data were analyzed using a combination of RTA and Bayesian techniques. First, the traditional log-log plot was produced to identify the linear flow production regime. Using the linear flow production data and traditional RTA equations, Bayesian inversion was carried out using two distinct Bayesian methods. The “rjags” and “EasyABC” packages in the open-source statistical software R were used for the traditional and approximate inversion, respectively. Model priors were based on (1) information available about the Marcellus Shale from technical literature and (2) results from a hydraulic fracturing forward model. Posterior distributions and credible intervals were produced for the fracture length, matrix permeability, and skin factor. These credible intervals were then compared with true reservoir and hydraulic fracturing data. The methodology was also repeated for an actual case in the Barnett shale. The most substantial finding was that for nearly all the investigated cases—including complicated scenarios (such as including finite fracture conductivity, fracturing fluid flowback, and heterogeneity in fracture length in the reservoir/hydraulic fracturing forward model)—the combined RTA-Bayesian model provided a 95% credible interval that encompassed the true values of the reservoir/hydraulic fracture parameters. We also found that the choice of the prior distribution did not affect the posterior distribution/credible interval in a significant manner as long as it was moderately concentrated and consistent with engineering science. Also, a comparison of the approximate Bayesian computation (ABC) and the traditional Bayesian algorithms showed that the ABC algorithm reduced computational time by at least an order of magnitude with minimal loss in accuracy. In addition, the production history used, the number of iterations, and the tolerance of fitting in the ABC analysis had a minimal impact on the posterior distribution after an optimal point, which were determined to be at least 1-year production history, 10,000 iterations, and 0.001, respectively. In summary, the RTA-Bayesian production analysis method was implemented using relatively user-friendly computational platforms [R and Excel® (Microsoft Corporation, Redmond, Washington, USA)]. This methodology provided reasonable characterization of all key variables such as matrix permeability, fracture length, and skin when compared to results obtained from analytical methods. This probabilistic characterization has the potential to enable better understanding of well performance ranges expected from shale gas wells. The methodology described here can also be generalized to shale oil systems during linear flow.

Doctor-patient communication and trust in doctors during COVID 19 times-A cross sectional study in Chennai, India.

The COVID 19 pandemic created a global public health crisis. Physical distancing, masks, personal protective equipment worn by the doctors created difficulties in effective doctor-patient communication. This study was conducted to assess the difficulties faced by patients in communicating with their doctors due to the COVID 19 preventive measures, and its impact on the trust on their doctors. A cross sectional study of 359 persons attending a tertiary care center in Chennai, sampled in a non-probabilistic manner selected from the outpatient department, wards, and isolation facilities, was conducted using a questionnaire containing items covering three dimensions namely difficulties faced in accessing the health facility, difficulties in doctor-patient communication and trust in the doctors. The data were collected using Google Forms and analyzed using GNU PSPP open-source statistical software version 1.4.0. More than 60% of the participants complained of difficulty in accessing the health facility. More than 60% had difficulties in communicating with the doctors. There was a high level of trust in doctors among more than 80% of the participants. Comparison of the mean scores revealed that accessibility was a problem across ages, sexes, education and occupation groups. Communication barriers decreased with age and increased with education, but trust increased with age, but reduced with increasing education. Multivariable linear regression analysis revealed that difficulties in communication had a negative impact on trust (β = -0.63, p<0.001) and increasing education had a negative impact on trust (β = -0.42, p = 0.034). The COVID 19 pandemic and the preventive strategies such as lock-down, physical distancing, face mask and personal protective equipment created barriers to effective doctor patient communication and led to some compromise in trust in doctors during this time.