Relative Importance Of Variables Research Articles

The snow cover is more important than other climatic variables on the prediction of vegetation dynamics in the Pyrenees (1981–2014)

The dynamics of the mountain vegetation is governed by multiple climatic drivers including temperature, precipitation, radiation and snow cover variability. However, in the Mediterranean environment, little is known about the relative importance of each variable. In this study we assess how different snowpack indices (the maximum annual accumulation, the length of the snow season, and the melt-out date) and key climate variables (precipitation, temperature and shortwave solar radiation) control the interannual variability of the maximum Normalized Difference Vegetation Index (peak NDVI) in the Pyrenees. We use a 33 year long remote sensing dataset (1981–2014) to build a statistical model relating the annual peak NDVI with snow and climate variables. In elevated areas characterized by a well developed seasonal snowpack the melt-out date was the most important climatic variable for predicting the annual peak NDVI. However, at lower elevations where snow presence is ephemeral, shortwave solar radiation was the most important variable. This change in the relative importance of climatic variables occurs around 1300 m a.s.l. The results do not show a significant contribution of maximum snow accumulation, suggesting that indicators of snow presence (i.e. melt-out date or snow season duration), which are significantly easier to obtain than snow mass indicators from remote sensing, could be used to model the influence of the snowpack on peak NDVI at regional scale.

Evaluating spray drift from Uncrewed Aerial Spray Systems: A machine learning and variance-based sensitivity analysis of environmental and spray system parameters

Uncrewed Aerial Spray Systems (UASS), commonly called drones, have become an important application technique for plant protection products in Asia and worldwide. As such, environmental variables and spray system parameters influencing spray drift deserve detailed investigations. This study presents the data analysis of 114 UASS drift trials conducted between December 2021 and December 2022 in China. Study design was based on the ISO 22866:2005 protocol for spray drift trials and considered different UASS platforms, nozzles, and release heights, and specifically continuously measured weather conditions. The relative importance of the environmental variables and spray system parameters was evaluated by a random forest (RF) feature importance analysis, a Sobol sensitivity analysis and partial dependence plots. This approach was preferred to linear ranking techniques such as ANOVA (analysis of variance) due to the non-linearity of the system. In addition, partial dependence plots are proposed to visualize the relationship between specific input parameters within the system.Drift deposition curves calculated from the 114 trials show good agreement with previous UASS trials reported in the literature. As reported in previous studies, spray drift following UASS applications is lower than for manned aerial vehicles, greater than for ground spray applications, and similar to drift observed from orchard air blast applications. In addition, 9 trials were conducted on corn fields in order to evaluate the potential effect of crop cover on spray drift. Spray drift was observed to be reduced over the cropped soil, suggesting that plant cover might possibly reduce spray drift. These findings could help supporting drift mitigation policies, stewardship advice and product labelling around the world.

Taxonomic distinctness and diversity patterns of a polychaete (Annelida) community on the continental shelf of the Southern Gulf of Mexico.

The spatial patterns of taxonomic diversity of annelid polychaete species from the continental shelf in the Southern Gulf of Mexico were examined in this study. We used taxonomic distinctness and its spatial variations to explore the diversity patterns and how they change between Southern Gulf of Mexico regions. In addition, using taxonomic distinctness as a dissimilarity measure and Ward's Clustering, we characterized three distinct faunal assemblages. We also investigated patterns of richness, taxonomic distinctness, and distance decay of similarity between sampling stations as a ß-diversity measure. Finally, we examined the spatial relationships between polychaete assemblages and environmental variables to test the relative importance of spatial and environmental components in annelid polychaete community structure from the Southern Gulf of Mexico. We used a combination of eigenvector-based multivariate analyses (dbMEMs) and distance-based redundancy analysis (dbRDA) to quantify the relative importance of these explanatory variables on the spatial variations of taxonomic distinctness. The significance level of spatial and environmental components to the distribution of polychaete species showed that the combined effect of spatial processes and sediment characteristics explained a higher percentage of the variance than those parameters could alone.

Cluster-based regression using variational inference and applications in financial forecasting

This paper describes an approach to simultaneously identify clusters and estimate cluster-specific regression parameters from the given data. Such an approach can be useful in learning the relationship between input and output when the regression parameters for estimating output are different in different regions of the input space. Variational Inference (VI), a machine learning approach to obtain posterior probability densities using optimization techniques, is used to identify clusters of explanatory variables and regression parameters for each cluster. From these results, one can obtain both the expected value and the full distribution of predicted output. Other advantages of the proposed approach include the elegant theoretical solution and clear interpretability of results. The proposed approach is well-suited for financial forecasting where markets have different regimes (or clusters) with different patterns and correlations of market changes in each regime. In financial applications, knowledge about such clusters can provide useful insights about portfolio performance and identify the relative importance of variables in different market regimes. An illustrative example of predicting one-day S&P change is considered to illustrate the approach and compare the performance of the proposed approach with standard regression without clusters. Due to the broad applicability of the problem, its elegant theoretical solution, and the computational efficiency of the proposed algorithm, the approach may be useful in a number of areas extending beyond the financial domain.

Using machine learning models to estimate Escherichia coli concentration in an irrigation pond from water quality and drone-based RGB imagery data

The rapid and efficient quantification of Escherichia coli concentrations is crucial for monitoring water quality. Remote sensing techniques and machine learning algorithms have been used to detect E. coli in water and estimate its concentrations. The application of these approaches, however, is challenged by limited sample availability and unbalanced water quality datasets. In this study, we estimated the E. coli concentration in an irrigation pond in Maryland, USA, during the summer season using demosaiced natural color (red, green, and blue: RGB) imagery in the visible and infrared spectral ranges, and a set of 14 water quality parameters. We did this by deploying four machine learning models – Random Forest (RF), Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGB), and K-nearest Neighbor (KNN) – under three data utilization scenarios: water quality parameters only, combined water quality and small unmanned aircraft system (sUAS)-based RGB data, and RGB data only. To select the training and test datasets, we applied two data-splitting methods: ordinary and quantile data splitting. These methods provided a constant splitting ratio in each decile of the E. coli concentration distribution. Quantile data splitting resulted in better model performance metrics and smaller differences between the metrics for both the training and testing datasets. When trained with quantile data splitting after hyperparameter optimization, models RF, GBM, and XGB had R2 values above 0.847 for the training dataset and above 0.689 for the test dataset. The combination of water quality and RGB imagery data resulted in a higher R2 value (>0.896) for the test dataset. Shapley additive explanations (SHAP) of the relative importance of variables revealed that the visible blue spectrum intensity and water temperature were the most influential parameters in the RF model. Demosaiced RGB imagery served as a useful predictor of E. coli concentration in the studied irrigation pond

Elevational Patterns of Forest Evapotranspiration and Its Sensitivity to Climatic Variation in Dryland Mountains

Elevational climatic heterogeneity, complex terrains, and varying subsurface properties affect the sensitivity of evapotranspiration (ET) in dryland mountain forests to hydrometeorological changes. However, the elevational distribution of ET sensitivity and its major influencing factors remain poorly understood. This study focused on the mid-altitude zone (1000–3500 m) forests in the Chinese Western Tianshan Mountains and assessed ET sensitivity to multiple climate variables, including precipitation (P) and potential evapotranspiration (PET), from 2000 to 2020. To evaluate the multi-year mean and trends in ET sensitivity, multi-source remote sensing data and regional survey data were analyzed using Spearman’s correlation coefficient, the sliding window method, and Kendall’s test. Furthermore, the relative importance of environmental variables (topography, geology, soil, and vegetation) was investigated. P and PET showed no significant trends, while ET exhibited a significant increasing trend (5.81 mm/yr, p < 0.01), particularly at elevations above 2000 m. Most forests (93.5%) showed a positive sensitivity of ET to P, and 70.0% showed a positive sensitivity of ET to PET, mainly at elevations of 1500–2500 m. Additionally, the trend in ET sensitivity to P decreased with an increasing elevation, with 64.5% showing a positive trend. Meanwhile, the trend in ET sensitivity to PET increased with elevation, with 88.1% showing a positive trend. Notably, 53.2% of the forests showed increasing ET sensitivity trends to both P and PET, primarily at elevations of 2000–3000 m with a mean normalized difference vegetation index (NDVI) of 0.56. Geological factors, particularly the hydrological properties of weathered bedrock, contributed the most (~47%) to mean sensitivity. However, geological and vegetative factors, including the NDVI and root zone water availability, were the main contributors (35% each) to the sensitivity. This study highlights the elevation-dependent sensitivity of dryland mountain forests to hydrothermal changes, with higher-elevation forests (>2000 m) being more sensitive to global warming.

Machine learning strategy for light lamb carcass classification using meat biomarkers

In Mediterranean areas, lamb meat is considered to be of great commercial value. Moreover, consumers are becoming increasingly interested in understanding the origin of lamb meat and its associated production and breeding systems. Among many applications, algorithms based on artificial intelligence are used to identify the origin of food products, and in this context, algorithms such as the Support Vector Machine (SVM), K-Nearest Neighbours (KNN), and the Artificial Neural Network (ANN) have been proposed to differentiate the origin of the animals according to their feeding diet. The objective of this study was to evaluate the performance of a variable reduction method based on a multiple regression model and three widely-used machine learning algorithms (SVM, KNN and ANN) for the classification of three commercial light lamb carcasses, from three feeding diets, in an indigenous Spanish breed (Mallorquina), using fatty acid and volatile compound biomarkers of meat. Machine learning algorithms were employed to discriminate lamb carcasses using 14 identified significant biomarkers, which were arranged based on an estimation of the relative importance (stepwise forward multiple regression F-score) of the input variables. We achieved high performances for the SVM, KNN and ANN algorithms, with 86%, 98% and 98% prediction accuracy, respectively. Among the 14 biomarkers used, 7 were identified as showing the highest discriminant capacity. The F-scores indicate that C17:1 and C20:5 n-3 fatty acids, and 2,5-dimethylpyrazine and 3-methylbutanal volatile compounds are the four most relevant biomarkers for predicting three lamb feeding diets.

Machine Learning-Based Causality Analysis of Human Resource Practices on Firm Performance

An organization’s human resource management practices are essential for its competitive advantage. This study specifically examined human resource (HR) practices that predict corporate performance (employee turnover and firm sales) based on a backpropagation neural network (BPN)-based causality analysis. This study aims to test how to optimize human resource practices to improve organizational performance. This study elucidated the effect of HR practices and organizational-level factors on predicting employee turnover and firm sales. The BPN-based causality analysis revealed the relative importance of explanatory variables on firm performance. To test the model, it employed the Human Capital Corporate Panel open data on Korean companies’ HR practices and other characteristics. The analysis identifies causal relationships between specific HR practices and firm performance. The results show that compensation-related HR practices are most influential in predicting firm sales and employee turnover. Moreover, training-related HR practices were modest, and talent acquisition and performance management practices had relatively weak effects on the two outcomes. The study provides insights into how human resource practices can be optimized to improve firm performance and enhance organizational effectiveness. The findings of this study contribute to the growing body of research on the use of machine learning in HR management and suggest practical implications for managers’ insights to optimize HR practices.

Data-driven prediction of dimensionless quantities for semi-infinite target penetration by integrating machine-learning and feature selection methods

This study employs a data-driven methodology that embeds the principle of dimensional invariance into an artificial neural network to automatically identify dominant dimensionless quantities in the penetration of rod projectiles into semi-infinite metal targets from experimental measurements. The derived mathematical expressions of dimensionless quantities are simplified by the examination of the exponent matrix and coupling relationships between feature variables. As a physics-based dimension reduction methodology, this way reduces high-dimensional parameter spaces to descriptions involving only a few physically interpretable dimensionless quantities in penetrating cases. Then the relative importance of various dimensionless feature variables on the penetration efficiencies for four impacting conditions is evaluated through feature selection engineering. The results indicate that the selected critical dimensionless feature variables by this synergistic method, without referring to the complex theoretical equations and aiding in the detailed knowledge of penetration mechanics, are in accordance with those reported in the reference. Lastly, the determined dimensionless quantities can be efficiently applied to conduct semi-empirical analysis for the specific penetrating case, and the reliability of regression functions is validated.

Variable importance and scale of influence across individual scottish wildcat hybrid habitat models

Understanding the scale dependence of species-habitat relationships is an important area of research in species distribution modeling. There has been little research focused on how habitat selection may depend on individual variation among organisms, geographical location and ecological context of that location. Furthermore, little is known about the extent and drivers of heterogeneity of scale dependence among individuals of a species inhabiting different ecological contexts, and few studies have compared scale dependence and variable importance in a spatially replicated framework. Two of the most important factors for interpreting habitat relationships models include: (1) the relative importance of variables in the model and (2) the spatial scale at which each variable has the largest influence. Based on the existing evidence we hypothesize a priori that landcover variables will generally be the most important predictors, followed by topography, then soil type (which influence both vegetation and prey), Normalized Difference Vegetation Index (NDVI) as an indicator of total vegetation density and perhaps a proxy for prey density, vegetation cover and rabbit abundance. We also expected a priori that there would be consistent patterns of scale dependence across individual wildcat hybrid models related to different variable groups. We expected topographical features to be selected at broad scales, as they influence broad-scale climatic and ecological conditions. We also expected that land cover classes and vegetation cover density to be selected at relatively broad scales given past research showing land cover generally influences habitat selection at relatively broad scales. We expected NDVI and soil type to be selected at finer scales, as their variation influences the distribution of resources and limiting conditions within landscapes. Finally, we expected that rabbit abundance and linear features would affect wildcat hybrid occurrence at the finest scales, given these are resources and conditions that vary over short distances and strongly influence wildcat and wildcat hybrid behavior and habitat use. Our results were consistent with the hypothesis that there may be consistency regarding which variables or variable groups are most important as predictors of wildcat hybrid occurrence in Scotland. Based on previous research we expected that there would be consistent patterns of scale dependence across individual wildcat hybrid models related to different variable groups. Finally, our results identify a clear and consistent trend of increasing frequency of inclusion of variables at increasingly broad scales. This is a linear trend in frequency of variables retained increasing as the scale increased. This suggests a consistent and monotonic pattern of more frequent retention of variables at increasingly broad scales.

Modeling net ecosystem exchange of CO2 with gated recurrent unit neural networks

The complex forcing mechanisms of biophysical drivers on ecosystem productivity, as well as some unknown relationships, stipulates the application of artificial intelligence techniques in modeling the ecosystem carbon cycle. In particular, various deep learning algorithms have been successfully applied to comprehend temporal relationships, though some issues regarding capturing temporal features for modeling CO2 fluxes across time scales remain unsolved. In this study, a gated recurrent unit (GRU) model was trained for simulating half-hourly net ecosystem exchange of CO2 (NEE) using 12-year continuous flux data (2009–2020) from seven experimental bioenergy crops in southwest Michigan, USA. The fields were historically managed either as corn-soybean rotation agricultural (AGR) or Conservation Reserve Program (CRP) lands and planted to no-till corn (AGR-C and CRP-C), restored-prairie (AGR-Pr and CPR-Pr), switchgrass (AGR-Sw and CRP-Sw) and smooth brome grass (CRP-Ref). We compared NEE simulations for the entire year, growing season, and non-growing season datasets, and analyzed relative importance of biophysical variables on variations of NEE. GRU performed well in simulating NEE at single site, with R2 of 0.89–0.93, 0.85–0.90, and 0.61–0.85 for the annual, growing season, and non-growing season datasets, respectively, which provided comparable accuracy and 6 % of run-time shorter than those of long short-term memory. R2 reached 0.70 and 0.80 across all sites and unique land-cover types. Radiation parameters contributed most to the variability of NEE. Contributions of individual variables appear more complex during the non-growing season and include incoming shortwave radiation, day of year, temperature, Monin-Obukhov stability, wind direction, and soil water content. The six most important forcing variables are consistent with our current understanding, although their ranked importance varied by ecosystem type and modeling scale. These results bring renewed insights in modeling observations at seasonal scales and provide guidance for variable selection to scale up measurements from one site to multiple land-cover types across the landscape.

Investigation and the development of learning analytics dashboard in open and distance learning using big data mining

<p class="Abstract">The main aim of this study is to provide universities with a way of examining and predicting student performance. The fundamental aim and purpose of this study is to help academic institutions to analyse and predict student performance. The credibility and accuracy of the model was examined by comparing the predicted results of the model with the observed values. And educational data mining techniques were used to create student profiles. Weighted gain, classification analysis, decision tree and rule induction were used in this study. The results of the study showed that the level of students' academic performance varied according to criteria such as academic structure, faculty, mode of enrolment and gender. In order to determine the relative importance of variables, the information weight gain technique was used after generating rule induction parameters and hidden rules between data. Using data mining techniques, we can obtain both guidelines to instruct students and information to help us identify them.</p>

Investigation of light gas adsorption by microporous materials using data analysis and decision tree machine learning algorithm

This study aimed to extract insights into the adsorption behavior of light gases by microporous materials such as zeolites, MOFs, and activated carbons. Data reported in 22 articles published between 1974 and 2022 were analyzed using a decision tree (DT) machine learning algorithm. A comprehensive database comprising 3297 data points, elucidating the impacts of 8 input variables on adsorption capacity, was constructed. Various exploratory data analysis techniques, including histograms, bar charts, and scatter plots, were employed to discern the relationships among input variables and the performance variable. Additionally, a DT model was employed to regress the adsorbed capacity data. Furthermore, a parametric study of this model facilitated the determination of the relative importance of input variables and their partial dependence, enhancing the interpretability of the model and enabling the deduction of heuristics for high or low adsorption capacity. The exploratory data analysis revealed that pressure and molecular mass of the adsorbate were the most significant variables influencing adsorption capacity.

Synoptic circulation factors associated with wintertime high-PM2.5 concentrations in seoul, Republic of Korea: Their interpretations and applications

Many studies have explored atmospheric circulation fields associated with high particulate matter (PM) events, but only a few studies have evaluated the relative importance of meteorological variables in the circulation fields. In this study, we applied factor analysis method to examine meteorological variables related to the synoptic pattern of high-PM2.5 episodes (i.e., days with daily averaged PM2.5 concentration exceeding 35 μg m−3) in Seoul, Republic of Korea, during the winter season of 2004–2018. Here, we highlight three independent key factors (F1, F2, and F3) associated with high-PM2.5 concentrations. F1 is related to temperature (T) and meridional wind (V) at 850 hPa and 1000 hPa (i.e., T850, T1000, V850, and V1000) and temperature and geopotential height (Z) at 500 hPa (i.e., T500 and Z500). It is the most strongly correlated with T850, indicating an increase in thermal stability over Korea during high-PM2.5 events due to vertically reduced atmospheric ventilation. Meanwhile, F2 is significantly related to the geopotential height and zonal wind (U) at 850 hPa and 1000 hPa (i.e., Z850, Z1000, U850, and U1000). This indicates a dynamically stable condition due to the anomalous easterlies caused by the vertically elongated anticyclone over the northern part of the Korean Peninsula. Finally, F3 is positively associated with zonal and meridional winds at 500 hPa (i.e., U500 and V500). This means that relatively strong winds in the low-to-mid troposphere contribute to high-PM2.5 concentrations by transporting pollutants from the industrial regions of eastern China to Korea. We also examined the possible influence of climate change on high-PM2.5 concentrations based on the aforementioned three factors. According to the various shared socioeconomic pathway (SSP) scenarios, F1 increases significantly for all SSP scenarios compared to F2 and F3, indicating that synoptic patterns related to F1 occur frequently, leading to a favorable condition for high-PM2.5 concentrations. Thus, our finding suggests that climate change alone could worsen air quality in the future even without changes in direct PM2.5 emissions.

On the rocks: Biogeography and floristic identity of rocky ecosystems in eastern South America

AbstractThe geodiversity of rocky ecosystems includes diverse plant communities with specific names, but their continental‐scale floristic identity and the knowledge on the role of macroclimate remain patchy. Here, we assessed the identity of plant communities in eastern Brazil across multiple types of rocky landscapes and evaluated the relative importance of climatic variables in constraining floristic differentiation. We provided lists of diagnostic species and an assessment of the conservation status of the identified floristic groups. We compiled a data set of 151 sites (4498 species) from rocky ecosystems, including campos rupestres, campos de altitude, granitic‐gneiss lowland inselbergs, and limestone outcrops. We used unsupervised clustering analysis followed by ANOSIM to assess floristic groups among sites. We performed a random forest variable selection to test whether the identified floristic groups occupy distinct climatic spaces. Six groups (lithobiomes) segregated floristically according to lithology and climate. Alongside campos de altitude and limestone outcrops, inselbergs were divided according to the biome in which they occur (Atlantic Forest or Caatinga), and campos rupestres were largely segregated according to their lithological matrix (ironstone or quartzitic). Plant communities of Caatinga inselbergs were more similar to limestone outcrops, while Atlantic Forest inselbergs communities resembled campos de altitude. The composition of plant communities on outcrops seems to be largely constrained by lithology, but climatic factors are also meaningful for sites with similar lithology. The current network of protected areas does not cover these unique ecosystems and their floristic heterogeneity, with Caatinga inselbergs and limestone outcrops being the least protected.

The Impact of Shared Instructional Leadership and Social Capital on School Effectiveness

The aim of this study is to determine the effect of shared instructional leadership and social capital on school effectiveness according to teacher perceptions. The study was designed in a descriptive relational survey model. The sample of the study consists of 403 teachers working in schools in the Odunpazarı and Tepebaşı districts located in the city center of Eskişehir, Turkey in the 2021-2022 academic year. In the study, “Shared instructional leadership scale,” “Social capital scale” and “School effectiveness scale” were used. In the study, it was found that teachers’ perceptions of shared instructional leadership, social capital, and school effectiveness levels were high. It was determined that shared instructional leadership and social capital variables together had a high and significant relationship with school effectiveness. The relative importance of the predictor variables on school effectiveness are shared instructional leadership and social capital variables, and these variables were found to be significant predictors of school effectiveness. School principals can contribute to the strengthening of social capital and school effectiveness by exhibiting shared instructional leadership behaviors such as focusing on student learning, sharing authority and responsibilities, strong dialogue and cooperation.

Analysis of Octane Retention Prediction Model for Catalytic Cracked Gasoline Based on Ridge Regression Model and Gradient Descent Optimization

On the basis of the given material, in order to increase the RON retention of the catalytic cracking unit, the prediction model of gasoline octane retention and the best operation variable inversion model were established based on the Ridge regression model and Gradient descent method. First, based on the Ridge regression model, the leave-one method is used to obtain the relative importance of the operational variables, and select the most important variables, so as to reduce the characteristic dimension of the model; Then, the RON retention prediction model is trained based on the Ridge regression model; Finally, based on the trained Ridge regression model and its weight parameters, the optimal operating variables were optimized separately using the gradient when the operation variable has a range or no range of value. The experimental results show that when 146 are selected from 361 operating variables, the model loss value stabilizes; when α is 0.6, the test set R2 is 0.9882, test set MSE is 0.0193, and the comprehensive performance is better than the random forest, support vector machine model; When the operation variable has two categories of value range and no value range, 2,000 times, the best inversion value of the operation variable makes the RON retention prediction value of the test sample similar to the expected value, and the MAE drops to 2.89999�10-3 and 7.62939�10-6, respectively. In conclusion, the RON retention prediction model proposed in this study has good results, and the best operating variable can be reversed, based on the given material parameters, making the optimal RON retention quantity.

Determinants of spring migration departure dates in a New World sparrow: Weather variables reign supreme.

Numerous factors influence the timing of spring migration in birds, yet the relative importance of intrinsic and extrinsic variables on migration initiation remains unclear. To test for interactions among weather, migration distance, parasitism, and physiology in determining spring departure date, we used the Dark-eyed Junco (Junco hyemalis) as a model migratory species known to harbor diverse and common haemosporidian parasites. Prior to spring migration departure from their wintering grounds in Indiana, USA, we quantified the intrinsic variables of fat, body condition (i.e., mass ~ tarsus residuals), physiological stress (i.e., ratio of heterophils to lymphocytes), cellular immunity (i.e., leukocyte composition and total count), migration distance (i.e., distance to the breeding grounds) using stable isotopes of hydrogen from feathers, and haemosporidian parasite intensity. We then attached nanotags to determine the timing of spring migration departure date using the Motus Wildlife Tracking System. We used additive Cox proportional hazard mixed models to test how risk of spring migratory departure was predicted by the combined intrinsic measures, along with meteorological predictors on the evening of departure (i.e., average wind speed and direction, relative humidity, and temperature). Model comparisons found that the best predictor of spring departure date was average nightly wind direction and a principal component combining relative humidity and temperature. Juncos were more likely to depart for spring migration on nights with largely southwestern winds and on warmer and drier evenings (relative to cooler and more humid evenings). Our results indicate that weather conditions at take-off are more critical to departure decisions than the measured physiological and parasitism variables.

Forecasting the Nexus and Impact of News Sentiment on NYSE, Gold Prices, and WTI Oil Using The Neural Network Approach

The study aims to determine the impact of news sentiment during COVID-19 and the nexus between the NYSE, Gold prices, and WTI oil prices. For this purpose, we use hourly data of news sentiment and commodity prices hourly data from February 10, 2020, hours 1000 to March 06, 2020 hours 1600 hours, with the help of a multilayer artificial neural network. Further, We use the connection weight approach to observe the relative importance of input variables that help predict output variables. The results suggest that all the indexes are interconnected, where the increase/decrease of any index impacts the other commodity index. Further, the news index remains the second highest predictor for gold price and WTI oil price indexes. In addition, the news index also seems to be an essential predictor of these commodities.
 Keywords: NYSE, WTI Oil, Gold, Covid-19, Neural Network, Connection Weight

Improving the estimation of atmospheric water vapor pressure using interpretable long short-term memory networks

Atmospheric water vapor pressure is an essential meteorological control on land surface and hydrologic processes. As it is not as frequently observed as other meteorologic conditions, it is often inferred through the August–Roche–Magnus formula by simply assuming dew point and daily minimum temperatures are equivalent or by empirically correlating the two temperatures using an aridity correction. The performance of both methods varies considerably across different regions and during different time periods; obtaining consistently accurate estimates across space and time remains a great challenge. Here, an interpretable Long Short-Term Memory (iLSTM) network conditioned on static, location specific attributes is proposed to estimate the daily vapor pressure. This approach allows for training a single transferable model using ensemble data from multiple sites and exploring the quantitative dependency of vapor pressure prediction on multiple environmental variables and their histories. To evaluate this approach, three iLSTM model configurations were developed, each considering different site attributes as static variables. For each configuration, multiple model realizations were trained using 83 FLUXNET sites in the United States and Canada, where each realization corresponds to different withheld groups of sites used for model evaluation. Results show that the iLSTM networks noticeably improve the estimation accuracy in comparison with the two assumption-based methods for most sites, reducing the failure rate from 32 % to 10.9 % for the best iLSTM model configuration. Additionally, this network provides reasonable insights into both the relative importance of the time-series input variables and their temporal importance. This method is found to be effective for imputing vapor pressure across space and time.