Influential Explanatory Variables Research Articles

Assessing physical and biological lake oxygen indicators using simulated environmental variables and machine learning algorithms

We integrate observations and simulated data from physics-based models with observations and machine learning (ML) algorithms to assess and predict lake dissolved oxygen (DO) and Apparent Oxygen Utilization (AOU). DO is a proxy of hypoxia, and AOU a proxy of respiration processes and biological activity. Weather, hydrology, and agroecosystem data were used to understand how various environmental drivers impact hypoxic conditions in Lake Erie for a 15-year period. We utilized outputs from various physics-based models and developed ML models with Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) to predict DO and AOU and rank the importance of influential explanatory variables. The ML models were able to accurately predict bottom DO and AOU, and identified thermal stratification as the most influential environmental variable, followed by mineralized phosphorus soil content. RF and XGBoost were not statistically different, therefore, we recommend the use of either ML algorithm to study hypoxic lake conditions.

Identification of Reasons for Culpable Homicides and Attempted Murders: A Case Study for the Kelaniya Police Division

Culpable Homicides and attempted murders are ultimate crimes that could create ripple effects on a society which could go far beyond the original loss of human life. Owing to the unpredictable nature of such crimes that require complex investigations the objective of this study was to come up with an appropriate model to identify the reason for a culpable homicide or an attempted murder using a statistical approach. This study use data collected from 12 Police stations in Kelaniya Police Division relating to the incidents happened between 2010 and 2020. The Pearson Chi-square test was used in identifying the influential explanatory variables. Out of the 18 variables, 8 predictors including Weapon used, Relationship, Location, Civil Status of the perpetrator were statistically associated with the identified reasons at 5 % level of significance. Multinomial logistic regression followed by four data mining models including classification tree, support vector machine (SVM), k-nearest neighbour (KNN), and probabilistic neural network (PNN) were employed initially with a training and testing set which was randomly selected in the ratio 90:10. The 4 data mining models were then fitted separately by using the bagging technique. The accuracies were compared using the confusion matrixes and rates of misclassifications of the critical classes. Out of the fitted models, the highest accuracy of 93.75 % was shown by the PNN model with a spread of 0.6. The identified model can be used as a decision support tool by crime investigators and relevant authorities for wise decision making.

Environmental and Economic Factors Affecting COVID-19 Infection in 175 Countries

This study explores a potential correlation between environmental and economic factors and the confirmed COVID-19 cases. This article uses official infection rate data and environmental and economic factors that may help accelerate transmissions, such as temperature, humidity, and GDP. Using statistical methods, i.e., ordinary least squares and quantile regression, our findings indicate that temperature, moisture, and island condition were the most influential explanatory variables. Nevertheless, the other factors, which are population density of a country, GDP, and country’s regime, are not statistically significant, indicating that those factors do not significantly explain changes in the confirmed COVID-19 cases.

COVID 19 Pandemic, Socio-Economic Behaviour and Infection Characteristics: An Inter-Country Predictive Study Using Deep Learning.

This work aims to develop a data driven multi-horizon incidence forecasting model considering the inter-country variability in static socio-economic factors. The specific objectives of this study are to predict the future country-wise COVID 19 incidences, to locate the influences of individual socio-economic factors on the predictions, to analyze the clusters of countries on the basis of influential explanatory variables and thus to search for intra-cluster and inter-cluster characteristics. To that respect this study has used the deep neural network based temporal fusion transformer for the predictions, Pearson correlation to understand the influence of socio-economic variables on incidence and hierarchical clustering for cluster-analysis. The findings conclude that the inter-country infection related predictions vary widely over spatio-temporal variability and different socio-economic variables have different influences over this inter-country variability. It is observed that greater the population size, stronger the global connectedness, larger the social cohesion, higher the population density and meaningful the gender based discrimination higher will be the future spread. On the other hand greater the development level, higher the nutritional status, greater the access to quality health services, greater the urban population and greater the material poverty lesser will be the future spread. Definite spatial pattern of influence of the explanatory variables emerged from cluster analysis. To minimize the vulnerability towards unforeseen biological calamities modern and sustainable development policies are needed; affluence may not guarantee less infection. But these policies should vary between economies due to the variation in socio-economic status of the countries worldwide.

An optimized surface aerodynamic temperature approach to estimate maize sensible heat flux and evapotranspiration

Typical errors for single-source sensible heat flux estimation models range from 15-40% for vegetated surfaces, with 10 to 20% errors introduced by atmosphere stability functions and remaining errors due to the quality of the input data and assumptions regarding the surface aerodynamic temperature (To). This paper describes the development and evaluation of an optimized To model, which serves as key input to estimate and improve sensible heat fluxes for maize fields. The best resulting optimized linear model included four variables: air temperature, surface temperature, fractional vegetation cover, and a proposed new variable that accounts for wind speed and the relative angle of attack between wind direction and crop row orientation. During the 2017 to 2019 maize growing seasons, on two different fields located at a Research Farm near Greeley, Colorado, USA, field data were collected. Micrometeorological data were measured in situ at the height of 3.30 m above the ground surface. Nadir-looking stationary infrared thermometers measured surface temperature. Multispectral surface reflectance data were collected on-site weekly using a handheld multispectral radiometer. Net radiation and soil heat fluxes were acquired at approximately ¼ and ½ of the field's length. Sensible heat fluxes were measured with two sets of Large Aperture Scintillometers to calibrate the To model. A one-way Analysis of Variance was performed to model To for different ranges of maize leaf area index values. A local sensitivity analysis was performed to identify the most influential explanatory variables in the To optimization model. The sensitivity analysis indicated that uncertainty on air temperature and surface temperature measurements were responsible for the most variability in sensible heat flux estimates when using the developed optimized To model. Overall, results indicated that the optimized To model improved the estimation of maize sensible heat fluxes by 31%; which resulted in an improvement on latent heat flux (or evapotranspiration) by 9% compared to a non-optimized linear To model. The interactions between the crop row layout and wind direction promote more turbulent mixing for heat transfer mechanisms.

Assessing the Role of Snow Cover for Post‐Wildfire Revegetation Across the Pacific Northwest

AbstractForested, mountain landscapes in the Pacific Northwest (PNW) are changing at an unprecedented rate, largely due to shifts in the regional climate regime. Documented climate warming trends across the PNW include increasing wildfire frequency and severity and an increasingly ephemeral snowpack, especially at moderate elevations. We analyzed 24 high severity wildfires across four distinct PNW mountainous subregions, examining snow‐vegetation relationships for two years pre‐fire and four years post‐fire. To assess the importance of snow cover for revegetation compared to other climatic, topographic, and burn severity‐related variables, binary regression tree models were constructed for the dominant pre‐fire conifer species within each of the four PNW subregions. Summer precipitation consistently appeared as the most important variable driving post‐fire revegetation across all four subregions. Snow cover variables (snow cover frequency and snow disappearance date), along with elevation, were shown to be secondary but significantly influential explanatory variables for revegetation in the Oregon and Washington Cascades. Revegetation was also analyzed using a time series of linear regressions across 200‐m elevation bands by measuring correlations between winter snow cover and summer vegetation greenness. Results showed strong positive post‐fire correlations at moderate elevations in the western Montana Rockies and at the lowest elevation band in the Idaho Rockies. Considering trends of increasing wildfire activity, lower snowpacks, and earlier snow disappearance dates across the PNW, forests will likely experience more frequent drought conditions that will impact post‐wildfire vegetation regrowth.

The Geography of the Covid-19 Pandemic: A Data-Driven Approach to Exploring Geographical Driving Forces.

The Covid-19 pandemic emerged and evolved so quickly that societies were not able to respond quickly enough, mainly due to the nature of the Covid-19 virus’ rate of spread and also the largely open societies that we live in. While we have been willingly moving towards open societies and reducing movement barriers, there is a need to be prepared for minimizing the openness of society on occasions such as large pandemics, which are low probability events with massive impacts. Certainly, similar to many phenomena, the Covid-19 pandemic has shown us its own geography presenting its emergence and evolving patterns as well as taking advantage of our geographical settings for escalating its spread. Hence, this study aims at presenting a data-driven approach for exploring the spatio-temporal patterns of the pandemic over a regional scale, i.e., Europe and a country scale, i.e., Denmark, and also what geographical variables potentially contribute to expediting its spread. We used official regional infection rates, points of interest, temperature and air pollution data for monitoring the pandemic’s spread across Europe and also applied geospatial methods such as spatial autocorrelation and space-time autocorrelation to extract relevant indicators that could explain the dynamics of the pandemic. Furthermore, we applied statistical methods, e.g., ordinary least squares, geographically weighted regression, as well as machine learning methods, e.g., random forest for exploring the potential correlation between the chosen underlying factors and the pandemic spread. Our findings indicate that population density, amenities such as cafes and bars, and pollution levels are the most influential explanatory variables while pollution levels can be explicitly used to monitor lockdown measures and infection rates at country level. The choice of data and methods used in this study along with the achieved results and presented discussions can empower health authorities and decision makers with an interactive decision support tool, which can be useful for imposing geographically varying lockdowns and protectives measures using historical data.

Evaluation of long-term urban transitions in a megacity’s building sector based on alternative socioeconomic pathways

Climate change has been one of the central issues for long-term transformations of cities. However, environmental loads have not been effectively taken into account for future urban plans. Specifically, possible reductions of carbon emissions from cities significantly rely on technological progress and human lifestyle changes, among other factors, which may be a barrier to estimating future environmental loads for cities. There is, therefore, urgent need for developing methods to integrate climate change policies with urban sustainable development planning. This research attempts to estimate future carbon emissions in the building sector in the megacity of Tokyo, considering the social and urban form transitions. The alternative socioeconomic pathways on an urban scale have been developed by the authors. The development of quantitative analysis methods linked with socioeconomic scenario assumptions and finding relevant parameters, also required novel thinking. In this research, influential explanatory variables for building stock changes are investigated and applied for the scenario analysis. Final results show two alternative ‘sustainability’ pathways that can reduce carbon emissions by approximately 15 percent compared to the Business As Usual scenario in 2050. This result encourages the development of various innovative sustainable pathways that are the most suitable for each city.

Multiple Linear Regression Models for Predicting Nonpoint-Source Pollutant Discharge from a Highland Agricultural Region

Sediment runoff from dense highland field areas greatly affects the quality of downstream lakes and drinking water sources. In this study, multiple linear regression (MLR) models were built to predict diffuse pollutant discharge using the environmental parameters of a basin. Explanatory variables that influence the sediment and pollutant discharge can be identified with the model, and such research could play an important role in limiting sediment erosion in the dense highland field area. Pollutant load per event, event mean concentration (EMC), and pollutant load per area were estimated from stormwater survey data from the Lake Soyang basin. During the wet season, heavy rains cause large amounts of suspended sediment and the occurrence of such rains is increasing due to climate change. The explanatory variables used in the MLR models are the percentage of fields, subbasin area, and mean slope of subbasin as topographic parameters, and the number of preceding dry days, rainfall intensity, rainfall depth, and rainfall duration as rainfall parameters. In the MLR modeling process, four types of regression equations with and without log transformation of the explanatory and response variables were examined to identify the best performing regression model. The performance of the MLR models was evaluated using the coefficient of determination (R2), root mean square error (RMSE), coefficient of variation of the root mean square error (CV(RMSE)), the ratio of the RMSE to the standard deviation of the observed data (RSR) and the Nash–Sutcliffe model efficiency (NSE). The performance of the MLR models of pollutant load except total nitrogen (TN) was good under the condition of RSR, and satisfactory for the NSE and R2. In the EMC and load/area models, the performance for suspended solids (SS) and total phosphorus (TP) was good for the RSR, and satisfactory for the NSE and R2. The standardized coefficients for the models were analyzed to identify the influential explanatory variables in the models. In the final performance evaluation, the results of jackknife validation indicate that the MLR models are robust.

Do biological traits drive geographical patterns in European amphibians?

AbstractAimsThe present‐day biogeography of European amphibians has been hypothesized to have arisen from range expansion and recolonization of the northern part of the continent from southern late Pleistocene refugia, such that northern species generally possess large ranges while southerly species are mostly small‐ranged. Here we test the hypothesis that these patterns are likely to be underpinned by biological traits associated with dispersal ability. We do this by analysing data for anurans and urodeles, the two main groups of European amphibians.LocationEurope.MethodsWe built a database of biological traits (body size, fecundity, life span, habitat specialization) of European amphibians, excluding island endemics. We mapped the basic macroecological patterns of range size and position, and analysed the causal pathways for range size using structural equation models (SEMs).ResultsAmphibian species with a small range size are largely restricted to areas in southern Europe associated with putative Pleistocene refugia. Those present in northern Europe are exclusively large‐ranged species whose distributions extend all the way from southern Europe. SEMs explained 54% of range size variation for anurans, with long life span and high fecundity being influential explanatory variables, and explained 61% of range size variation within urodeles, with measures of species fecundity being influential.Main conclusionsSpecies that have successfully recolonized the north following deglaciation have the largest ranges for both groups of amphibians. These large‐ranged species generally possess traits that indicate the potential for rapid range expansion, with differences apparent in the balance of traits between anurans and urodeles. The traits linked to northern distributions (and large range size) appear to be a mix of r and K traits, indicating that intermediate life‐history strategies have proved to be optimal for range expansion into northern regions. These results integrate species biology with geographical history in explaining present‐day patterns of species distribution, range size and diversity.

Grazing intensity effects on soil quality: A spatial analysis of a Mediterranean grassland

Intensive grazing, such as that occurring in the vicinity of animals' shaded concentration points (CPs), causes the degradation of many soil properties, creating patterns as a function of distance from the CPs. The aim of this research is to characterize the spatial patterns of land degradation as a function of grazing intensity, in terms of soil quality, with respect to (1) the distance from CPs; (2) grazing density; and (3) additional environmental variables, in typical Mediterranean grassland, in the eastern Galilee region of Israel. To fulfill this aim, a soil quality index (SQI) was developed based on the integration of 12 physical, biological, and chemical soil properties. Soil properties and quality surface maps were developed by using the spline interpolation technique as a geostatistical analysis on a small spatial scale. In addition, a spatial statistics technique, namely the geographically weighted regression (GWR), was used to measure the effect of different explanatory variables (grazing density, distance from CPs, vegetation cover, aspect, slope, and landscape cover classification) on the SQI. The results show that near the CPs, there is a clear effect of grazing activity on all soil properties and nutrients that can be attributed to a low SQI. Significantly lower soil quality was found around the CPs with moderate and high stocking rates than in the pasture and the control area, where grazing was excluded. Applying the GWR model, using the landscape-cover classification as the explanatory variable for the SQI, provided the best results in the spatial analysis with R2adjusted=0.97. The second and third most influential explanatory variables for the SQI were the distance from the CPs and grazing density, with R2adjusted=0.94 and R2adjusted=0.93, respectively. This outcome demonstrates grazing intensity effects on SQ and its complexity, mainly due to the impact of landscape cover and grazing density. In conclusion, the presented framework was found adequate and could potentially be made operational for assessing SQ under different grazing intensities and degradation stages in a spatial scale.

Impact of temporal resolution of inputs on hydrological model performance: An analysis based on 2400 flood events

Summary Hydro-climatic data at short time steps are considered essential to model the rainfall–runoff relationship, especially for short-duration hydrological events, typically flash floods. Also, using fine time step information may be beneficial when using or analysing model outputs at larger aggregated time scales. However, the actual gain in prediction efficiency using short time-step data is not well understood or quantified. In this paper, we investigate the extent to which the performance of hydrological modelling is improved by short time-step data, using a large set of 240 French catchments, for which 2400 flood events were selected. Six-minute rain gauge data were available and the GR4 rainfall–runoff model was run with precipitation inputs at eight different time steps ranging from 6 min to 1 day. Then model outputs were aggregated at seven different reference time scales ranging from sub-hourly to daily for a comparative evaluation of simulations at different target time steps. Three classes of model performance behaviour were found for the 240 test catchments: (i) significant improvement of performance with shorter time steps; (ii) performance insensitivity to the modelling time step; (iii) performance degradation as the time step becomes shorter. The differences between these groups were analysed based on a number of catchment and event characteristics. A statistical test highlighted the most influential explanatory variables for model performance evolution at different time steps, including flow auto-correlation, flood and storm duration, flood hydrograph peakedness, rainfall–runoff lag time and precipitation temporal variability.

Analysing pseudoephedrine/methamphetamine policy options in Australia using multi-criteria decision modelling

BackgroundIn this paper we capture and synthesize the unique knowledge of experts so that choices regarding policy measures to address methamphetamine consumption and dependency in Australia can be strengthened. We examine perceptions of the: (1) influence of underlying factors that impact on the methamphetamine problem; (2) importance of various models of intervention that have the potential to affect the success of policies; and (3) efficacy of alternative pseudoephedrine policy options. MethodsWe adopt a multi-criteria decision model to unpack factors that affect decisions made by experts and examine potential variations on weight/preference among groups. Seventy experts from five groups (i.e. academia (18.6%), government and policy (27.1%), health (18.6%), pharmaceutical (17.1%) and police (18.6%)) in Australia participated in the survey. ResultsSocial characteristics are considered the most important underlying factor, prevention the most effective strategy and Project STOP the most preferred policy option with respect to reducing methamphetamine consumption and dependency in Australia. One-way repeated ANOVAs indicate a statistically significant difference with regards to the influence of underlying factors (F(2.3, 144.5)=11.256, p<.001), effectiveness of interventions (F(2.4, 153.1)=28.738, p<.001) and policy options (F(2.8, 175.5)=70.854, p<.001). ConclusionA majority of respondents believed that genetic, biological, emotional, cognitive and social factors are the most influential explanatory variables in terms of methamphetamine consumption and dependency. Most experts support the use of preventative mechanisms to inhibit drug initiation and delayed drug uptake. Compared to other policies, Project STOP (which aims to disrupt the initial diversion of pseudoephedrine) appears to be a more preferable preventative mechanism to control the production and subsequent sale and use of methamphetamine. This regulatory civil law lever engages third parties in controlling drug-related crime. The literature supports third-party partnerships as it engages experts who have knowledge and expertise with respect to prevention and harm minimization.

Cross-scale contradictions in ecological relationships

Not accounting for spatial heterogeneity in ecological analyses can cause modeled relationships to vary across spatial scales, specifically different levels of spatial resolution. These varying results hinder both the utility of data collected at one spatial scale for analyses at others and the determination of underlying processes. To briefly review existing methods for analyzing data collected at multiple scales, highlight the effects of spatial heterogeneity on the utility of these methods, and to illustrate a practical statistical method to account for the sources of spatial heterogeneity when they are unknown. Using simulated examples, we show how not accounting for the drivers of spatial heterogeneity in statistical models can cause contradictory findings regarding relationship direction across spatial scales. We then show how mixed effects models can remedy this multiscaling issue. Ignoring sources of spatial heterogeneity in statistical models with coarse spatial scales produced contradictory results to the true underlying relationship. Treating drivers of spatial heterogeneity as random effects in a mixed effects model, however, allowed us to uncover this true relationship. Mixed effects models is advantageous as it is not always necessary to know the influential explanatory variables that cause spatial heterogeneity and no additional data are required. Furthermore, this approach is well documented, can be applied to data having various distribution types, and is easily executable using multiple statistical packages.

Development of multiple linear regression models as predictive tools for fecal indicator concentrations in a stretch of the lower Lahn River, Germany

Since rivers are typically subject to rapid changes in microbiological water quality, tools are needed to allow timely water quality assessment.A promising approach is the application of predictive models. In our study, we developed multiple linear regression (MLR) models in order to predict the abundance of the fecal indicator organisms Escherichia coli (EC), intestinal enterococci (IE) and somatic coliphages (SC) in the Lahn River, Germany. The models were developed on the basis of an extensive set of environmental parameters collected during a 12-months monitoring period. Two models were developed for each type of indicator:1) an extended model including the maximum number of variables significantly explaining variations in indicator abundance and 2) a simplified model reduced to the three most influential explanatory variables, thus obtaining a model which is less resource-intensive with regard to required data.Both approaches have the ability to model multiple sites within one river stretch. The three most important predictive variables in the optimized models for the bacterial indicators were NH4–N, turbidity and global solar irradiance, whereas chlorophyll a content, discharge and NH4–N were reliable model variables for somatic coliphages.Depending on indicator type, the extended mode models also included the additional variables rainfall, O2 content, pH and chlorophyll a.The extended mode models could explain 69% (EC), 74% (IE) and 72% (SC) of the observed variance in fecal indicator concentrations. The optimized models explained the observed variance in fecal indicator concentrations to 65% (EC), 70% (IE) and 68% (SC). Site-specific efficiencies ranged up to 82% (EC) and 81% (IE, SC). Our results suggest that MLR models are a promising tool for a timely water quality assessment in the Lahn area.

The Effects of Capital on Interprovincial Migration: A Nova Scotia Focused Assessment

This paper examines the issue of interprovincial migration and its potential relationship to capital expenditure. This paper estimates and reviews the relationship between several variables and their effects on interprovincial migration; one of its main contributions is to analyze the effect of capital expenditure on net interprovincial migration. This issue was examined from a Nova Scotia perspective through both literature and an empirical model. A time-series cross sectional regression model found both median income and GDP per capita to be significant and influential explanatory variables. Capital expenditure was found to be statistically significant, but negatively related and negligible. Based on these findings and the literature review, four recommendations are proposed: continue with the status quo, incentivize return migration, provide a tax break for out of province commuters, and establish a system to expedite accreditation.

Unpacking performance benefits of innovation ambidexterity: Evidence from the pharmaceutical industry

1. IntroductionInnovation, or new technology-based products development in particular, is one of the most influential explanatory variables for favorable organizational performance in our age. However, this does not mean that we can explain organizational performance by solely considering innovation enabled by findings of revolutionarily new technological knowledge (henceforth, we refer to such innovation as exploratory innovation). Conversely, innovation enabled by new applications of existing technological knowledge (henceforth, exploitative innovation) also influences organizational performance substantially. For example, customers expect firms to innovate incrementally by exploiting existing technological knowledge, rather than by exploring novel technological knowledge (Christensen & Bower, 1996). Firms may also choose to forego opportunities to explore new product technology because their suppliers and distributors support exploitation of existing product technology (Glasmeier, 1991). Accordingly, the importance of simultaneously pursuing exploratory as well as exploitative innovation is underscored by recent research on organizational ambidexterity (O'Reilly & Tushman, 2008, among others).Prior work on organizational ambidexterity uncovers performance benefits enabled by organizational ambidexterity with respect to various organizational aspects, including product configuration (Salvador, Chandrasekaran, & Sohail, 2014), modes of operation (Stettner & Lavie, 2014), innovation strategy (Cao, Gedajlovic, & Zhang, 2009; Derbyshire, 2014; He & Wong, 2004; Lubatkin, Simsek, Ling, & Veiga, 2006), search behaviors (Wang & Li, 2008), firm or business-unit contexts (Chang, Yang, & Chen, 2009; De Clercq, Thongpapanl, & Dimov, 2013; Gibson & Birkinshaw, 2004; Patel, Messersmith, & Lepak, 2013), organizational design (Adler, Goldoftas, & Levine, 1999; Chang et al., 2009), as well as founding team's prior affiliations (Beckman, 2006).However, the prior research neglects to explicitly distinguish two distinct mechanisms in which organizational ambidexterity enables favorable organizational performance as it does not separately examine differences between two alternative complementary relationships between exploitation and exploration (Gupta, Smith, & Shalley, 2006; Katila & Ahuja, 2002). We refer to the two complementary relationships as a mutually compensatory relationship and a mutually enabling relationship, respectively. The former refers to a relationship in which exploitation and exploration mutually compensate for each other's deficiencies, in that they contribute to organizational performance in distinct, but mutually complementary ways. Put differently, exploitation's effects on organizational performance complement exploration's effects, and vice versa. On the other hand, in a mutually enabling relationship, exploitation and exploration mutually seeds and primes each other. More precisely, acts of exploitation complement acts of exploration by enabling the latter, and vice versa.It is of theoretical as well as practical importance to take into account the differences between these alternative complementary relationships because they offer distinct explanations on the way in which ambidexterity enables favorable organizational performance. In addition to the locus of complementarity (i.e., effects or acts) and the mechanisms of performance benefits (i.e., compensatory or enabling), the two alternative complementary relationships also differ in their influences on the negative externalities between exploitation and exploration (Boumgarden, Nickerson, & Zenger, 2012). Furthermore, their differences inform us two distinct research directions for pursuing ambidexterity antecedents.In this manuscript, we aim to fill this research gap by unpacking two alternative complementary relationships between exploitation and exploration, thereby arguing that organizational ambidexterity with respect to product innovation is positively associated with organizational performance in two distinct mechanisms. …

Study on relationships between climate-related covariates and pipe bursts using evolutionary-based modelling

Researchers extensively studied external loads since they are widely recognized as significant contributors to water pipe failures. Physical phenomena that affect pipe bursts, such as pipe-environment interactions, are very complex and only partially understood. This paper analyses the possible link between pipe bursts and climate-related factors. Many water utilities observed consistent occurrence of peaks in pipe bursts in some periods of the year, during winter or summer. The paper investigates the relationships between climate data (i.e., temperature and precipitation-related covariates) and pipe bursts recorded during a 24-year period in Scarborough (Ontario, Canada). The Evolutionary Polynomial Regression modelling paradigm is used here. This approach is broader than statistical modelling, implementing a multi-modelling approach, where a multi-objective genetic algorithm is used to get optimal models in terms of parsimony of mathematical expressions vs. fitting to data. The analyses yielded interesting results, in particular for cold seasons, where the discerned models show good accuracy and the most influential explanatory variables are clearly identified. The models discerned for warm seasons show lower accuracy, possibly implying that the overall phenomena that underlay the generation of pipe bursts during warm seasons cannot be thoroughly explained by the available climate-related covariates.

A joint model of destination and mode choice for urban trips: a disaggregate approach

Trip destination and mode choice are highly influenced by travelers' perceptions and behaviors; selecting a destination and a vehicle for a trip are two interdependent problems. This paper presents and applies a disaggregate joint model for traveler destination and mode choice. The choice model uses fuzzy set and probability theory to deal with the uncertainty embedded in travelers' perceptions and behaviors. The model is structured as a decision tree in which the fuzzy and non-fuzzy classification of influential variables regarding destination selection and mode choice expand the tree. The most influential explanatory variables among all the variables categorized for travelers' household, trip, and living zone specifications are selected based on the maximizing information. An aggregation method is designed to provide aggregate estimates for transportation planning based on the suggested disaggregate choice model. A data-set of over 9000 home-based morning peak-hour trips in Shiraz, a large city in Iran, is used for model construction and evaluation. When compared with a multinomial logit (MNL) model, the suggested models' estimates are more accurate than the traditional MNL model.

Likelihood-Based Boosting in Binary and Ordinal Random Effects Models

This article presents a likelihood-based boosting approach for fitting binary and ordinal mixed models. In contrast to common procedures, this approach can be used in high-dimensional settings where a large number of potentially influential explanatory variables are available. Constructed as a componentwise boosting method, it is able to perform variable selection with the complexity of the resulting estimator being determined by information criteria. The method is investigated in simulation studies both for cumulative and sequential models and is illustrated by using real datasets. The supplementary materials for the article are available online.