Explore the vast range of titles available.

Machine learning (ML) was used to assess and predict urban air temperature (Tair) considering the complexity of the terrain features in Yerevan (Armenia). The estimation was performed based on the Partial Least-Squares Regression (PLSR) model with a high number (30) of input variables. The relevant parameters include a newly purposed modification of spectral index IBI-SAVI, which turned out to strongly impact Tair prediction together with land surface temperature (LST). Cross-validation analysis on temperature predictions across a station-centered 1000 m circular area revealed quite a high correlation (R2Val = 0.77, RMSEVal = 1.58) between the predicted and measured Tair from the test set. It was concluded the remote sensing is an effective tool to estimate Tair distribution where a dense network of weather stations is not available. However, further developments will include incorporation of additional weather parameters from the weather stations, such as precipitation and wind speed, as well as the use of non-parametric ML techniques.

Not all research reports explicitly address this question. Nevertheless, this approach is especially helpful when studies involve complex experimental designs incorporating multiple independent variables or sophisticated regression or structural equation models. For example, economists often specify stochastic equations corresponding to the “steps” in logistic multiple regression or time-series analyses used in economic forecasting.

We investigate the problem of testing whether d possibly multivariate random variables, which may or may not be continuous, are jointly (or mutually) independent. Our method builds on ideas of the two-variable Hilbert–Schmidt independence criterion but allows for an arbitrary number of variables. We embed the joint distribution and the product of the marginals in a reproducing kernel Hilbert space and define the d-variable Hilbert–Schmidt independence criterion dHSIC as the squared distance between the embeddings. In the population case, the value of dHSIC is 0 if and only if the d variables are jointly independent, as long as the kernel is characteristic. On the basis of an empirical estimate of dHSIC, we investigate three non-parametric hypothesis tests: a permutation test, a bootstrap analogue and a procedure based on a gamma approximation. We apply non-parametric independence testing to a problem in causal discovery and illustrate the new methods on simulated and real data sets.

Latent class analysis is a clustering method that is nowadays widely used in social science research. Researchers applying latent class analysis will typically not only construct a typology based on a set of observed variables but also investigate how the encountered clusters are related to other, external variables. Although it is possible to incorporate such external variables into the latent class model itself, researchers usually prefer using a three-step approach. This is the approach wherein after establishing the latent class model for clustering (step 1), one obtains predictions for the class membership scores (step 2) and subsequently uses these predicted scores to assess the relationship between class membership and other variables (step 3). Bolck, Croon, and Hagenaars (2004) showed that this approach leads to severely downward-biased estimates of the strength of the relationships studied in step 3. These authors and later also Vermunt (2010) developed methods to correct for this bias. In the current study, we extended these correction methods to situations where class membership is not predicted but used as an explanatory variable in the third step, a situation widely encountered in social science applications. A simulation study tested the performance of the proposed correction methods, and their practical use was illustrated with real data examples. The results showed that also when the latent class variable is used as a predictor of external variables, the uncorrected three-step approach leads to severely biased estimates. The proposed correction methods perform well under conditions encountered in practice.

Multiple myeloma (MM) is a disease of malignant plasma cells (PC) with poor survival. Disease progression and treatment relapse are attributed to MM cancer stem cells (CSCs) and signaling molecules such as MUC1 and XBP1. The study aimed to determine the prognostic value of expression of CSC-associated biomarkers, MUC1 and XBP1 in MM, which has not been explored previously. In this study, we determined the immunohistochemical expression of CSC markers (ALDH1, CD117, and CD34), MUC1, and XBP1 in 128 MM formalin-fixed paraffin-embedded bone marrow archival blocks. The expression of biomarkers was assessed for association with clinicopathological variables and patient survival. Descriptive analysis, survival plots and crude association between outcome and independent variables were assessed using Kaplan Meier and Log rank test. Univariate and multivariable analyses were performed using simple and multiple Cox regression models. The results are reported as crude and adjusted hazard ratios with 95% confidence intervals. Expression of ALDH1 and CD117 was found in 51% and 48% of the tumors, respectively. ALDH1 expression was associated with 1.83 years of reduced survival for patients with CD56-negative tumors. MUC1 expression was observed in 62%, whereas XBP1 was expressed in 48% of tumors. Combinatorial group analysis of XBP1 and MUC1 stratified patients into two prognostic groups. Cases with tumors negative for expression of MUC1 and XBP1 (XBP1-/ MUC1-) were categorized as a good prognostic group with increased survival of 3.42 years compared to cases with tumors expressing both ( Worst prognosis , XBP1 + /MUC1+) . Concordant expression of MUC1 and XBP1 in MM defines a subset of patients with adverse outcomes. The adjusted hazard ratio showed a four-fold increased risk of mortality associated with the concordant expression of MUC1 and XBP1 in patients > 65 years of age.

Estimating structural models is often viewed as computationally difficult, an impression partly due to a focus on the nested fixed-point (NFXP) approach. We propose a new constrained optimization approach for structural estimation. We show that our approach and the NFXP algorithm solve the same estimation problem, and yield the same estimates. Computationally, our approach can have speed advantages because we do not repeatedly solve the structural equation at each guess of structural parameters. Monte Carlo experiments on the canonical Zurcher bus-repair model demonstrate that the constrained optimization approach can be significantly faster.

Numerous papers in the fields of marketing and consumer behavior that utilize moderated multiple regression express concerns regarding the existence of a multicollinearity problem in their analyses. In most cases, however, as we show in this paper, the perceived multicollinearity problem is merely an illusion that arises from misinterpreting high correlations between independent variables and interaction terms.

Surface water quality parameters are important means for determination of water’s suitability for irrigation. In this research, data from 32 irrigation stations were used to calculate the sodium adsorption rate (SAR), sodium percentage (Na%), Kelly index (KI), permeability index (PI) and irrigation water quality index (IWQI) for evaluation of surface water quality. The obtained SAR, KI and Na% values, respectively, varied between 0.10 and 9.43, 0.03–1.37 meq/l and 3.16–57.82%. The calculated PI values indicate that, 93.75% of the water samples is in “suitable” category, and 6.25% is in “non-suitable” category. The IWQI values obtained from the research area varied between 30.59 and 81.09. In terms of irrigation water quality, 12.5% of the samples is of “good” quality, 15.62% is of “poor” quality, 68.75% is of “very poor” quality, and 3.12% is of “non-suitable” quality. Accordingly, IWQI value was estimated on the basis of SAR, Na%, KI and PI values using multiple regression and artificial neural network (ANN) model. The regression coefficient (R2) was determined as 0.6 in multiple regression analysis, and a moderately significant relationship (p < 0.05) was detected. As the calculated F value was higher than the tabulated F value, a real relationship between the dependent and independent variables is inferred. Four different models were built with ANN, and the statistical performance of the models was determined using statistical parameters such as average value (µ), standard error (SE), standard deviation (σ), R2, root mean square error (RMSE) and mean absolute percentage error (MAPE). The training R2 value belonging to the best model was found to be significantly high (0.99). The relation between the estimation results of ANN model and the experimental data (R2 = 0.92) verifies the model’s success. As a result, ANN proved to be a successful means for IWQI estimation using different water quality parameters.

This paper investigates the effectiveness of a blended learning environment through analyzing the relationship between student characteristics/background, design features and learning outcomes. It is aimed at determining the significant predictors of blended learning effectiveness taking student characteristics/background and design features as independent variables and learning outcomes as dependent variables. A survey was administered to 238 respondents to gather data on student characteristics/background, design features and learning outcomes. The final semester evaluation results were used as a measure for performance as an outcome. We applied the online self regulatory learning questionnaire for data on learner self regulation, the intrinsic motivation inventory for data on intrinsic motivation and other self-developed instruments for measuring the other constructs. Multiple regression analysis results showed that blended learning design features (technology quality, online tools and face-to-face support) and student characteristics (attitudes and self-regulation) predicted student satisfaction as an outcome. The results indicate that some of the student characteristics/backgrounds and design features are significant predictors for student learning outcomes in blended learning.

Resting-state functional magnetic resonance imaging (rs-fMRI) has been widely utilized to investigate plasticity mechanisms and functional reorganization in multiple sclerosis (MS). Among many resting state (RS) networks, a significant role is played by the salience network (SN, ventral attention network). Previous reports have demonstrated the involvement of osteopontin (OPN) in the pathogenesis of MS, which acts as a proinflammatory cytokine ultimately leading to neurodegeneration. Concentration of serum OPN was related to MRI findings 10.22±2.84 years later in 44 patients with MS. Local and interhemispheric correlations (LCOR, IHC), ROI-to-ROI and seed-based connectivity analyses were performed using serum OPN levels as independent variable along with age and gender as nuisance variables. We found significant associations between OPN levels and local correlation in right and left clusters encompassing the central opercular- and insular cortices (p-FDR = 0.0018 and p-FDR = 0.0205, respectively). Moreover, a significant association was identified between OPN concentration and interhemispheric correlation between central opercular- and insular cortices (p-FDR = 0.00015). Significant positive associations were found between OPN concentration and functional connectivity (FC) within the SN (FC strength between the anterior insula ventral division and 3 other insular regions, F(2,13) = 7.84, p-FDR = 0.0117). Seed-based connectivity analysis using the seven nodes of the SN resulted in several positive and inverse associations with OPN level. Serum OPN level may predict FC alterations within the SN in 10 years.