Multivariate Modeling Approach Research Articles

A Multivariate Local Rational Modeling Approach for Detection of Structural Changes in Test Vehicles

Abstract A data driven structural change detection method is described and evaluated where the data are acceleration and force measurements from a mechanical structure in the form of a vehicle. By grouping the measured signals as inputs and outputs an hypothesized MIMO linear dynamic relation between the inputs and outputs is assumed. It is assumed that baseline data are available to build statistical models for the estimated frequency function of the baseline system at selected frequencies. When new data is available, the monitoring algorithm re-estimates the non-parametric frequency function and uses a test statistic based on the statistical distance to detect possible change. To generate the frequency function estimates a non-parametric MIMO frequency function estimator based on the local rational model (LRM) method is developed. A statistical analysis of the proposed test statistic shows that it has an F-distribution for data from the baseline case. The method is evaluated on simulated data from a high fidelity full scale vehicle simulation generating both baseline data and data from a structurally changed vehicle. In the evaluation, the frequency response functions were estimated by the non-parametric LRM method, the parametric ARX estimate and the non-parametric ETFE. The results show that all three methods can detect the structural change while the LRM method is more robust with respect to the selection of the hyperparameters.

The Role of Cash-Waqf Capital Ventures to Supply Shock Liquidity of Sharia Banks in East Java: Multivariate Threshold Autoregressive Simulation Model

Adequacy of liquidity is very important for Islamic banks to carry out their main functions as Islamic financial intermediaries and provide financing for the real sector. However, many Islamic banks are proven to have a shock in providing liquidity due to the weak capital structure of the bank. While cash-waqf funds can be a potential source of capital for sharia banks if managed as capital investments in sharia banks with a mudharabah or musyrakah contract. Therefore, using the autoregressive multivariate threshold simulation model approach, this study attempts to analyze the dynamic interaction between the monetary policy regime, cash-capital capital ventures, Islamic bank liquidity and inflation in East Java. The participation of waqf funds on Islamic bank capital is simulated at 1%, 5% and 10% with two regimens of threshold models, namely a tightening monetary policy regime and a monetary policy easing regime. The data in this study were obtained from Statistik Perbankan Syariah (SFS) published by the OJK and BI with the observation period of 2014.M1 - 2018.M6. The results show that the inclusion of waqf funds in Islamic bank capital is able to reduce the length of the period and the sharia bank liquidity shock both in high interest rate policy regime and low interest rate regime.

Multivariate Bayesian hierarchical Gaussian copula modeling of the non-stationary traffic conflict extremes for crash estimation

Recent studies have demonstrated that single conflict indicators represent only a fractional aspect of the severity of a traffic interaction. As such, integrating several conflict indicators in a unified model can improve conflict-based crash estimation. This study develops a multivariate Bayesian hierarchical Gaussian copula modeling approach, which comprises a multivariate Gaussian copula and a Bayesian hierarchical structure. The former has generalized extreme value marginals and captures multivariate dependence among several conflict indicators, while the latter combines traffic conflicts from different sites, incorporating several covariates and site-specific unobserved heterogeneity. The copula approach offers the flexibility in modeling multivariate structures including the selection of margins from various parametric families of univariate distribution and the construction of parametric copulas which satisfies different types of dependence structure. A model estimation approach for the multivariate Bayesian hierarchical Gaussian copula model is proposed and applied to estimate rear-end crashes from four signalized intersections in the city of Surrey, British Columbia. The modified time to collision (MTTC), post encroachment time (PET), and deceleration rate to avoid a crash (DRAC) were employed as conflict indicators. Three covariates including traffic volume, shock wave area, and platoon ratio were considered to account for non-stationarity in conflict extremes. The results show that in terms of crash estimation accuracy, the multivariate Bayesian hierarchical Gaussian copula model outperforms both the bivariate Bayesian hierarchical Gaussian copula models and the recently proposed multivariate Bayesian hierarchical model in which multiple traffic conflict indicators were combined by one dependence parameter.

IMIX: a multivariate mixture model approach to association analysis through multi-omics data integration.

Integrative genomic analysis is a powerful tool used to study the biological mechanisms underlying a complex disease or trait across multiplatform high-dimensional data, such as DNA methylation, copy number variation and gene expression. It is common to perform large-scale genome-wide association analysis of an outcome for each data type separately and combine the results ad hoc, leading to loss of statistical power and uncontrolled overall false discovery rate (FDR). We propose a multivariate mixture model (IMIX) framework that integrates multiple types of genomic data and allows modeling of inter-data-type correlations. We investigated the across-data-type FDR control in IMIX and demonstrated lower misclassification rates at controlled overall FDR than established individual data type analysis strategies, such as the Benjamini-Hochberg FDR control, the q-value and the local FDR control by extensive simulations. IMIX features statistically principled model selection, FDR control and computational efficiency. Applications to The Cancer Genome Atlas data provided novel multi-omics insights into the genes and mechanisms associated with the luminal and basal subtypes of bladder cancer and the prognosis of pancreatic cancer. We have implemented our method in R package 'IMIX' available at https://github.com/ziqiaow/IMIX, as well as CRAN soon. Supplementary data are available at Bioinformatics online.

Forecasting future dental health expenditures: Development of a framework using data from 32 OECD countries.

ObjectiveTo (1) develop a framework for forecasting future dental expenditures, using currently available information, and (2) identify relevant research and data gaps such that dental expenditure predictions can continuously be improved in the future.MethodsOur analyses focused on 32 OECD countries. Dependent on the number of predictors, we employed dynamic univariate and multivariate modelling approaches with various model specifications. For univariate modelling, an auto‐regressive (AR) dynamic model was employed to incorporate historical trends in dental expenditures. Multivariate modelling took account of historical trends, as well as of relationships between dental expenditures, dental morbidity, economic growth in terms of gross domestic product and demographic changes.ResultsEstimates of dental expenditures varied substantially across different model specifications. Models relying on dental morbidity as one of the predictors performed worst regardless of their specification. Using the best‐fitted model specification, that is the univariate second‐order autoregression [AR(2)], the forecasted dental expenditures across 32 OECD countries amounted to US$316bn (95% forecasted interval, FI: 258‐387) in 2020, US$434bn (95%FI: 354‐532) in 2030 and US$594bn (95%FI: 485‐728) in 2040. Per capita spending in 2040 was forecasted to be highest in Germany (US$889, 95%FI: 726‐1090) and lowest in Mexico (US$52, 95%FI: 42‐64).ConclusionsThe present study demonstrates the feasibility and challenges in predicting dental expenditures and can serve as a basis for improvement towards more sustainable and resilient health policy and resource planning. Within the limitations of available data sources, our findings suggest that dental expenditures in OECD countries could increase substantially over the next two decades and vary considerably across countries. For more accurate estimation and a better understanding of determinants of dental expenditures, more comprehensive data on dental spending and dental morbidity are urgently needed.

Collective orientation and its implications for coordination and team performance in interdependent work contexts

Purpose This paper aims to explore the influence of collective orientation (CO) on coordination and team performance for interdependently working teams while controlling for person-related and team variables. Design/methodology/approach A total of 58 two-person-teams participated in a simulation-based firefighting task. The laboratory study took 2 h for each team. The effects of CO in tasks of increasing complexity were investigated under the consideration of control variables, and the relations between CO, coordination and team performance were assessed using a multivariate latent growth curve modeling approach and by estimating indirect effects in simple mediation models. Findings Team members high on CO performed significantly better than low-scoring members. The effect of CO on team performance was independent from an increasing task complexity, whereas the effect of CO on coordination was not. The effect of CO on team performance was mediated by coordination within the team, and the positive relation between CO and performance persists when including group efficacy into the model. Research limitations/implications As CO is a modifiable person-related variable and important for effective team processes, additional research on factors influencing this attitude during work is assumed to be valuable. Practical implications CO is especially important for highly interdependently working teams in high-risk-organizations such as the fire service or nuclear power plants, where errors lead to severe consequences for human beings or the environment. Originality/value No other studies showed the importance of CO for coordination and team performance while considering teamwork-relevant variables and the interdependence of work.

Evaluation of a multivariate analysis modeling approach identifying sources and patterns of nonpoint fecal pollution in a mixed use watershed

Surface waters listed on impaired waters (303 d) lists due to pathogen contamination pose a significant environmental and public health burden. The need to address impairments through the Total Maximum Daily Load (TMDL) process has resulted in method developments that successfully identify nonpoint fecal pollution sources by maximizing available resources to improve water quality. However, the ability of those methods to effectively and universally identify sources of fecal pollution requires further evaluation. The objective of this research was to assess the usefulness of a previously described multivariate statistical approach to identify common patterns influencing fate and transport of fecal pollutants from sources to receiving streams using the Tuckasegee River watershed in Western North Carolina as a test watershed. Two streams were routinely monitored using a targeted sampling approach to assess fecal pollution extent and identify nonpoint sources using canonical correlation and canonical discriminant analyses. Fecal pollution in the watershed varied spatially and temporally with significantly higher fecal coliform concentrations observed in Scott Creek (f = 9.49, p = 0.002) and during the summer months (f = 14.8, p < 0.0001). Canonical correlations described 62–67% of water quality variability and indicate that fecal pollution in portions of the watershed are influenced by stormwater runoff and fecal indicator bacteria resuspension from sediment, while fecal pollution in other portions are influenced by soil erosion and surface runoff. Canonical discriminant analyses indicate that LULC significantly influences the nature and extent of fecal pollution. These results demonstrate that chemical parameters are useful predictors of fecal pollution and can help identify nonpoint fecal pollution sources in relation to land use patterns and land management practices. This approach to water quality monitoring program design and data analysis may effectively and efficiently identify parameters that best predict fecal pollution to aid in development and implementation of effective TMDLs to remediate impaired waters.

Multivariate volatility forecasts for stock market indices

Volatility forecasts aim to measure future risk and they are key inputs for financial analysis. In this study, we forecast the realized variance as an observable measure of volatility for several major international stock market indices and accounted for the different predictive information present in jump, continuous, and option-implied variance components. We allowed for volatility spillovers in different stock markets by using a multivariate modeling approach. We used heterogeneous autoregressive (HAR)-type models to obtain the forecasts. Based an out-of-sample forecast study, we show that: (i) including option-implied variances in the HAR model substantially improves the forecast accuracy, (ii) lasso-based lag selection methods do not outperform the parsimonious day-week-month lag structure of the HAR model, and (iii) cross-market spillover effects embedded in the multivariate HAR model have long-term forecasting power.

A wearable micro-fuel cell sensor for the determination of blood alcohol content (BAC): a multivariate regression model approach

A proton exchange membrane (PEM) fuel cell-based wearable alcohol sensing device has been explored for the monitoring of blood alcohol content (BAC) in real time. The watch-like alcohol monitoring device detects alcohol from the perspiration of the human skin. This transcutaneous (transdermal) fuel cell sensor can detect a wide range of concentrations of alcohol, including a physiological range of concentration (5–100 mg/dl), with the detection time of ~ 5 s. A calibration model, such as principal component regression (PCR), was implemented to reduce standard errors and refine the calibration of this sensor by offsetting noise. The BAC values from the transcutaneous sensor were closer to the theoretical values, with 77.47% improvement in accuracy compared to readings from off the shelf breathalyzer. The results from human studies demonstrate that this fuel cell sensing device in accompany with the proposed PCR model can be employed for the determination of BAC through transcutaneous measurements.

Multivariate Bayesian hierarchical modeling of the non-stationary traffic conflict extremes for crash estimation

Recent studies have developed univariate and bivariate Bayesian hierarchical extreme value models to improve traffic conflict-based crash estimation through addressing the issues of non-stationarity of conflict extremes and individual conflict indicators representing only partial severity aspects of a traffic event. Although the use of bivariate models showed considerable improvements compared to univariate models, two traffic conflict indicators may not be sufficient to cover all severity aspects of a traffic event. Therefore, a high dimensional multivariate Bayesian hierarchical extreme value model based on more than two conflict indicators merits further research. This study develops a multivariate Bayesian hierarchical extreme value modeling approach, which comprises a multivariate extreme value model and a Bayesian hierarchical structure. The former integrates several traffic conflict indicators in a unified framework, while the latter combines traffic conflicts from different sites, incorporating several covariates and site-specific unobserved heterogeneity. A model estimation approach for the multivariate Bayesian hierarchical extreme value model is proposed and applied to estimate rear-end crashes from four signalized intersections in the city of Surrey, British Columbia. The modified time to collision (MTTC), post encroachment time (PET), and deceleration rate to avoid a crash (DRAC) were employed as conflict indicators. Three covariates including traffic volume, shock wave area, and platoon ratio were considered to account for non-stationarity in conflict extremes. The results show that in terms of crash estimation accuracy, the high dimensional multivariate (e.g., trivariate) Bayesian hierarchical extreme value model outperforms both the low dimensional multivariate (e.g., bivariate) and univariate Bayesian hierarchical extreme value models. This is likely due to the multivariate model having the capability of sharing information across several different conflict indicators.

Analysis of Transformed Upstream Bioprocess Data Provides Insights into Biological System Variation.

In recent years, multivariate data analysis (MVDA) and modeling approaches have found increasing applications for upstream bioprocess studies (e.g., monitoring, development, optimization, scale-up, etc.). Many of these studies look at variations in the concentrations of metabolites and cell-based measurements. However, these measures are subject to system inherent variations (e.g., changes in metabolic activity) but also intentional operational changes. It is proposed to perform MVDA and modeling on data representative of the underlying biological system operation, that is, the specific rates, which are per se independent of the scale, operational strategy (e.g., batch, fed-batch), and biomass content. Two industrial case studies are highlighted to showcase the approach: one is HEK medium performance comparison study and the other is CHO scale-up/-down study. It is shown that analyzing processes in this way reveals insights into behavior of the underlying biological system, which cannot to the same degree be deducted from the analysis of concentrations.

Modified Carroll’s pyramid of corporate social responsibility to enhance organizational performance of SMEs industry

Research questions/issuesThe study seeks to examine Carrolls’ pyramid model in small-medium enterprises (SMEs) as an effective business strategy for organizational performance enhancement in industry of developing country which is in the initial phase of Industry 4.0 advance. The Corporate social responsibility (CSR) pyramid is important tool for fostering environmentally-sustainable manufacturing by having green manufacturing processes, green supply chain management, and green products and holds that firms should engage in relevant decisions, actions, policies and practices that simultaneously fulfill all component parts: ethical, legal, economic, philanthropic and environmental. Research findings/insightsOrganizational innovation as a mediator and corporate image as a moderator were incorporated in the modified conceptual model of Carroll’s pyramid. 888 valid responses from Small medium enterprises (SMEs) professionals in Pakistan manufacturing sector were collected, and then the structural equation modeling (SEM) based multivariate and conditional process modeling approaches were employed to conduct the investigation. Results demonstrated that the dimensions of modified Carroll’s pyramid of CSR have an affirmative and positive impact on the organizational performance of industrial SMEs if they were properly addressed in their innovation strategies. The findings further concluded that organizational innovation and corporate image are the significant mediator and moderator respectively between the dimensions of Carroll’s pyramid of CSR and the organizational performance that further enhance the organization performance and productivity in terms of ethical and sustainable business development and overall bottom line in the light of industry 4.0 approach. The outcomes confirmed that the inclusion of Industry 4.0, Cleaner Production, and Circular Economy concepts in the SME sector provides a synergistic business opportunity that intent to payback towards sustainable development by improving environmental management, production efficiency, and socially sustainable development for SMEs. Theoretical/academic implicationsThis study provides empirical support for the modified Carol’s pyramid in industrial SMEs following Industry 4.0 approach. As such, it suggests new avenues of research for both the Industry 4.0 literature, as well as for the studies in CSR and organizational performance perspective. Practitioner/policy implicationsOutcomes of this study suggest the senior managers of SME in industrial sector that implementing the modified CSR model would be an innovative and competitive business strategy in industry, which can effectively improve the organizational performance in terms of ethically, ecologically, and economically through organizational innovations and allows to foster Industry 4.0 development.

DNA methylation and incident cardiovascular disease.

DNA methylation is an epigenetic mark that reflects both genetic and environmental influences over the life course and has the potential to be a robust biomarker for cardiovascular disease (CVD) risk. However, standard association studies linking DNA methylation and CVD are susceptible to reverse causation and may not directly translate into useful biomarkers of future disease. Studies of incident CVD represent a crucial tool for improving this evidence base. Recent investigations have started to provide links between DNA methylation and incident CVD. Epigenome-wide association studies have suggested individual genetic loci in which differential methylation exists prior to disease onset, while multivariate predictive modeling approaches have made progress towards realizing the potential for DNA methylation as a predictive biomarker of CVD risk. Meanwhile, complementary analysis strategies such as Mendelian randomization have provided clues as to the causality of these epigenomic associations. Taken together, this wave of studies provides the basis for a better understanding of CVD pathophysiology and the development of more confident biomarkers of CVD risk.

The spatial structure of socioeconomic disadvantage: a Bayesian multivariate spatial factor analysis

ABSTRACT Neighborhood socioeconomic disadvantage is a measure of socio-spatial inequality that has been shown to be associated with a variety of social, economic, and health outcomes. Existing studies that explore the local patterning of disadvantage often construct composite indices that summarize the interactions between multiple dimensions of social status, but do not consider if, and how, disadvantage exhibits spatial structure. This study applies a Bayesian multivariate factor analytic modeling approach to examine the spatial structure of socioeconomic disadvantage in Toronto, Canada. Socioeconomic disadvantage is modeled as an area-based composite index associated with three variables measuring low income, low-educational attainment, and low occupational status, and a series of models with different assumptions regarding the spatial structure of disadvantage are compared. The best-fitting model shows that the prevalence of low-income households has the strongest positive association with disadvantage and that spatial clustering is three times more important than spatial heterogeneity for explaining the spatial structure of disadvantage. The implications of this study for analyzing multivariate spatial data and for understanding the interactions amongst multiple dimensions of disadvantage are discussed.

Geographical verification of Slovenian milk using stable isotope ratio, multi-element and multivariate modelling approaches

This work examines the use of stable isotopes and elemental composition for determining geographical origin and authenticity of cow milk from four geographical regions of Slovenian. Samples (277) were collected during summer and winter (2012–2014). It was possible to discriminate milk samples according to the year, season and production region using discriminant analysis (DA). The overall temporal prediction variability was 84.6% and 56.4% for regional differences. It was also possible to discriminate milk from three geographic regions, although Alpine samples overlap with Dinaric and Pannonian ones. Prediction ability was the highest for the Pannonian (82.1%) and lowest (26.9%) for the Alpine region. Pairwise comparison using OPLS-DA also displaying good regional predictability (≥0.77) with δ13Ccas values and Br content carrying the most variance. A model based on DD-SIMCA was also developed and applied to the control of Slovenian milk. The results revealed the mislabeling of three Slovenian milk products.

Multivariate count data models for adoption of new transport modes in an organization-based context

This study investigates the potential market demand of shared-ride taxi and shuttle services designed to serve members of organizations in dense urbanized areas. It develops and compares two different multivariate count data modeling approaches, the multinomial distribution and the full enumeration of count alternatives, under an integrated choice and latent variable framework. The study accounts for day-to-day variability in commuting behavior, also known as multimodality, by modeling the weekly frequency of commuting by different travel modes instead of modeling choices for a single trip/day. Using stated preference data collected in the Spring of Academic Year 2016–2017, the models are applied to a case study of students who are highly dependent on private cars at the American University of Beirut (AUB), Lebanon. Policy analysis is conducted to investigate the impact of different price levels and modal attributes on the students’ mode choice behavior. Under practical scenarios, results show that more than 55% of students would adopt a multimodal travel behavior in a given week and that 9–20% of trips are expected to be made by shared-taxi and 12–25% by shuttle. Thus, modeling single trip/day choices instead of weekly decisions would lead to limitations in model forecasts related to the full impact of the proposed policies over longer periods. Results also show that the full enumeration model guarantees higher prediction accuracy and results in an estimate of value of time that is closer to other local estimates for the study area.

Comovement and Instability in Cryptocurrency Markets

We analyze the extent of comovement between daily price returns of nine major cryptocurrencies during the first three main phases of their development, from April 2013 to November 2018. We assess its evolution using bivariate and multivariate modeling approaches and detect pronounced time variation. Generally, comovement is initially low and positive, but increases between early 2017 and late 2018. We then adopt a right-tail version of the Augmented Dickey-Fuller unit root test to identify periods of mildly explosive behavior (statistical instability) in the Network Value to Transactions (NVT) ratio (a measure of the dollar value of cryptocurrency transaction activity relative to network value) of six cryptocurrencies. We show evidence of significant mild explosiveness in all of them. At the end of 2017 and in 2018, several cryptocurrencies experience often simultaneous instability associated with rising NVT ratios. Instability is a dominant feature of these markets.

Intra-annual fluctuations dominating temporal dynamics of benthic diatom assemblages in a Chinese mountainous river

Understanding temporal dynamics of community may provide insights on biological responses under environmental changes. However, our knowledge on temporal dynamics of river organisms is still limited. In the present study, we employed a multivariate time-series modeling approach with a long-term dataset (i.e. 72 consecutive months) to investigate temporal dynamics of benthic diatom communities in four sites located in a Chinese mountainous river network. We hypothesized that: (1) there are multi-scale temporal dynamics within the diatom community; (2) intra-annual fluctuations dominate the community dynamics; (3) diatom species composing the community respond distinctly to environmental changes. We found that intra-annual fluctuations with periodicities &lt;12 months explained 8.1–16.1% of community variation. In contrast, fluctuations with periodicities of 13–36 months and 37–72 months only accounted for 1.1–5.9% and 2.8–9.7% of variance in diatom community dynamics, respectively. Taxa correlating significantly to each significant RDA axis (namely, RDA taxa group) displayed distinct temporal dynamics. Conductivity, total nitrogen, and pH were important to most RDA taxa groups across the four sites while their effects were group-specific. We concluded that intra-annual dynamics dominated temporal variation in diatom communities due to community responses to local environmental fluctuations. We suggest that long-term monitoring data are valuable for identifying multiple-scale temporal dynamics within biological communities.

Do we need multivariate modeling approaches to model crash frequency by crash types? A panel mixed approach to modeling crash frequency by crash types

In safety literature, simulation-based multivariate framework is the most commonly employed approach for analyzing multiple crash frequency dependent variables. The current research effort contributes to literature on crash frequency analysis by suggesting an alternative and mathematically simpler approach for analyzing multiple crash frequency variables for the same study unit. The proposed recasts a multivariate distributional problem as a repeated measure univariate problem. Specifically, we employed a simpler panel random parameter based univariate model framework to analyze zonal level crash counts for different crash types. The empirical analysis is based on the traffic analysis zone (TAZ) level crash count data for both motorized and non-motorized crashes from Central Florida for the year 2016. The performance of the proposed framework is compared with the performance of the random parameter multivariate negative binomial model (RPMNB) using a host of metrics for estimation sample and hold-out sample. The resulting goodness of fit and predictive measures clearly highlight the comparable performance offered by the proposed framework relative to the commonly used RPMNB model with substantially fewer parameters. The comparison exercise is augmented by computing aggregate level elasticity effects for both PMNB and RPMNB models. The results clearly highlight the comparable performance offered by the proposed PMNB model relative to the traditional RPMNB model. In summary, the proposed framework allows for a parsimonious specification without compromising the model explanatory power and provides similar performance as the most traditional multivariate NB model for analyzing different crash dimensions.

Patterns of floral morphology in relation to climate and floral visitors.

The diversity of floral morphology among plant species has long captured the interest of biologists and led to the development of a number of explanatory theories. Floral morphology varies substantially within species, and the mechanisms maintaining this diversity are diverse. One possibility is that spatial variation in the pollinator fauna drives the evolution of spatially divergent floral ecotypes adapted to the local suite of pollinators. Another possibility is that geographic variation in the abiotic environment and local climatic conditions favours different floral morphologies in different regions. Although both possibilities have been shown to explain floral variation in some cases, they have rarely been competed against one another using data collected from large spatial scales. In this study, we assess floral variation in relation to climate and floral visitors in four oil-reward-specialized pollination interactions. We used a combination of large-scale plant and pollinator samplings, morphological measures and climatic data. We analysed the data using spatial approaches, as well as traditional multivariate and structural equation modelling approaches. Our results indicate that the four species have different levels of specialization, and that this can be explained by their climatic niche breadth. In addition, our results show that, at least for some species, floral morphology can be explained by the identity of floral visitors, with climate having only an indirect effect. Our results demonstrate that, even in very specialized interactions, both biotic and abiotic variables can explain a substantial amount of intraspecific variation in floral morphology.