Search results for: multivariate logistic regression

Authors: Nattabhorn Leamcharaskul

Abstract:

This study aims to examine determining factors of the dynamics of poverty in Thailand by using panel data of 3,567 households in 2007-2017. Four techniques of estimation are employed to analyze the situation of poverty across households and time periods: the multinomial logit model, the sequential logit model, the quantile regression model, and the difference in difference model. Households are categorized based on their experiences into 5 groups, namely chronically poor, falling into poverty, re-entering into poverty, exiting from poverty and never poor households. Estimation results emphasize the effects of demographic and socioeconomic factors as well as unexpected events on the economic status of a household. It is found that remittances have positive impact on household’s economic status in that they are likely to lower the probability of falling into poverty or trapping in poverty while they tend to increase the probability of exiting from poverty. In addition, not only receiving a secondary source of household income can raise the probability of being a never poor household, but it also significantly increases household income per capita of the chronically poor and falling into poverty households. Public work programs are recommended as an important tool to relieve household financial burden and uncertainty and thus consequently increase a chance for households to escape from poverty.

Keywords: difference in difference, dynamic, multinomial logit model, panel data, poverty, quantile regression, remittance, sequential logit model, Thailand, transfer

Procedia PDF Downloads 112

Authors: Girlly Marchlina Listyono, Abdurrokhim Abdurrokhim, Emi Sukiyah, Pulung Arya Pranantya

Abstract:

Jakarta Area is composed by deposit which has various lithology characteristics. Based on its lithology types, colors, textures, mineral dan organic content from 22 wells scattered on Jakarta, lithofacies analysis and intra-wells data correlation can be done. From the analysis, it can be interpretated that Jakarta deposit deposited in marine, transition and terrestrial depositional environments. Terrestrial deposit characterized by domination of relatively coarse clastics and content of remaining roots, woods, plants, high content of quartz, lithic fragment, calcareous and oxidated appearace. The thickness of terrestrial deposit is thickening to south. Transitional deposit characterized by fine to medium clastics with dark color, high content of organic matter, various thickness in any ways. Marine deposit characterized by finer clastics, contain remain of shells, fosil, coral, limestone fragments, glauconites, calcareous. Marine deposit relatively thickening to north. Those lateral variety caused by tectonic, subsidence and stratigraphic condition. Deposition of Jakarta deposit from the data research was started on marine depositional environment which surrounded by the event of cycle of regression and transgression then ended with regression which ongoing until form shore line in north Jakarta nowadays.

Keywords: deposit, Indonesia, Jakarta, sediment, stratigraphy

Procedia PDF Downloads 253

Authors: Dalson Figueiredo, Enivaldo Rocha, Ranulfo Paranhos, José Alexandre

Abstract:

This paper analyzes the effects of institutional capacity on corruption. Methodologically, the research design combines both descriptive and multivariate statistics to examine two original datasets based on secondary data. In particular, we employ a principal component model to estimate an indicator of institutional capacity for both state audit institutions and subnational judiciary courts. Then, we estimate the effect of institutional capacity on two dependent variables: (1) incidence of administrative irregularities and (2) time elapsed to judge corruption cases. The preliminary results using ordinary least squares, negative binomial and Tobit models suggest the same conclusions: higher the institutional audit capacity, higher is the probability of detecting a corruption case. On the other hand, higher the institutional capacity of state judiciary, the lower is the time to judge corruption cases.

Keywords: institutional capacity, corruption, state level institutions, evidence from Brazil

Procedia PDF Downloads 369

Authors: Abadou Yacine, Faid Hayette

Abstract:

Significant investigations were performed on the use and impact on physical properties along with the mechanical strength of the recycled and reused E-glass waste powder. However, it has been modelled how recycled display e-waste glass may affect the characteristics and qualities of dune sand mortar. To be involved in this field, an investigation has been done with the substitution of dune sand for recycled E-glass waste and constant water-cement ratios. The linear relationship between the dune sand mortar and E-glass mortar mix % contributes to the model's reliability. The experimental data was exposed to regression analysis using JMP Statistics software. The regression model with one predictor presented the general form of the equation for the prediction of the five properties' characteristics of dune sand mortar from the substitution ratio of E-waste glass and curing age. The results illustrate that curing a long-term process produced an E-glass waste mortar specimen with the highest compressive strength of 68 MPa in the laboratory environment. Anova analysis indicated that the curing at long-term has the utmost importance on the sorptivity level and ultrasonic pulse velocity loss. Furthermore, the E-glass waste powder percentage has the utmost importance on the compressive strength and improvement in dynamic elasticity modulus. Besides, a significant enhancement of radiation-shielding applications.

Keywords: ANOVA analysis, E-glass waste, durability and sustainability, radiation-shielding

Procedia PDF Downloads 57

Authors: Ng'etich K. Stephen

Abstract:

Abstract—Globally analytics in basketball has advanced tremendously in the last decade. Organizations are leveraging the insights to improve team and player performance and, in the long run, generate revenue out of it. Due to limited basketball game-related statistics in African competitions, teams are unaware of how they compete with other continental basketball teams. The purpose of this study is to evaluate the regional difference in basketball game-related statistics between African teams that played in the newly formed league, the basketball African league and the European league. The basketball African league, a competition created through the partnership between NBA and FIBA, offers a good starting point since it has valuable basketball metrics to analyze. This study sought to use multivariate linear discriminant analysis to identify the game-related statistics that discriminate the teams in Euro league and the basketball African league.

Keywords: basketball africa league, basketball, euroleague, fiba, africa

Procedia PDF Downloads 99

Authors: Neha Gupta, Sonika Verma, Seema Puri, Nikhil Tandon, Narendra K. Arora

Abstract:

The concurrent increasing prevalence of underweight and overweight/obesity among children with changing lifestyle and the rapid transitioning society has necessitated the need for a unifying/multi-level approach to understand the determinants of the problem. The present community-based cross-sectional research study was conducted to assess the associations between lifestyle behavior and food environment of the child at household, neighborhood, and school with the BMI of children (6-12 year old) (n=612) residing in three rural clusters of Palwal district, Haryana. The study used innovative and robust methods for assessing the lifestyle and various components of food environment in the study. The three rural clusters selected for the study were located at three different locations according to their access to highways in the SOMAARTH surveillance site. These clusters were significantly different from each other in terms of their socio-demographic and socio-economic profile, living conditions, environmental hygiene, health seeking behavior and retail density. Despite of being different, the quality of living conditions and environmental hygiene was poor across three clusters. The children had higher intakes of dietary energy and sugars; one-fifth share of the energy being derived from unhealthy foods, engagement in high levels of physical activity and significantly different food environment at home, neighborhood and school level. However, despite having a high energy intake, 22.5% of the recruited children were thin/severe thin, and 3% were overweight/obese as per their BMI-for-age categories. The analysis was done using multi-variate logistic regression at three-tier hierarchy including individual, household and community level. The factors significantly explained the variability in governing the risk of getting thin/severe thin among children in rural area (p-value: 0.0001; Adjusted R2: 0.156) included age (>10years) (OR: 2.1; 95% CI: 1.0-4.4), the interaction between minority category and poor SES of the household (OR: 4.4; 95% CI: 1.6-12.1), availability of sweets (OR: 0.9; 95% CI: 0.8-0.99) and cereals (OR: 0.9; 95% CI: 0.8-1.0) in the household and poor street condition (proxy indicator of the hygiene and cleanliness in the neighborhood) (OR: 0.3; 95% CI: 0.1-1.1). The homogeneity of other factors at neighborhood and school level food environment diluted the heterogeneity in the lifestyles and home environment of the recruited children and their households. However, it is evident that when various individual factors interplay at multiple levels amplifies the risk of undernutrition in a rural community. Conclusion: These rural areas in Haryana are undergoing developmental, economic and societal transition. In correspondence, no improvements in the nutritional status of children have happened. Easy access to the unhealthy foods has become a paradox.

Keywords: transition, food environment, lifestyle, undernutrition, overnutrition

Procedia PDF Downloads 180

Authors: Bing-Bing E. Goh

Abstract:

Plant counting is a labour intensive and time-consuming task for the farmers. However, it is an important indicator for farmers to make decisions on subsequent field management. This study is to evaluate the potential of Sentinel-2 images using statistical analysis to retrieve information on plant density for monitoring, especially during critical period at the beginning of March. The model was calibrated with in-situ data from 19 winter wheat fields in Republic of Ireland during the crop growing season in 2019-2020. The model for plant density resulted in R2 = 0.77, RMSECV = 103 and NRMSE = 14%. This study has shown the potential of using Sentinel-2 to estimate plant density and quantify plant establishment to effectively monitor crop progress and to ensure proper field management.

Keywords: winter wheat, remote sensing, crop monitoring, multivariate analysis

Procedia PDF Downloads 159

Authors: Samuel N. Forjuoh, Marcia G. Ory, Jaewoong Won, Samuel D. Towne, Suojin Wang, Chanam Lee

Abstract:

The public health burden of obesity is well established as is the influence of physical activity (PA) on the health and wellness of individuals who are obese. This study examined the influence of selected demographic, health, and socioenvironmental factors on the walking behaviors of middle-aged and older overweight and obese adults. Online and paper surveys were administered to community-dwelling overweight and obese adults aged ≥ 50 years residing in four cities in central Texas and seen by a family physician in the primary care clinic from October 2013 to June 2014. Descriptive statistics were used to characterize participants’ anthropometric and demographic data as well as their health conditions and walking, socioenvironmental, and more broadly defined PA behaviors. Then Pearson chi-square tests were used to assess differences between participants who reported walking the recommended ≥ 150 minutes for any purpose in a typical week as a proxy to meeting the U.S. Centers for Disease Control and Prevention’s PA guidelines and those who did not. Finally, logistic regression was used to predict walking the recommended ≥ 150 minutes for any purpose, controlling for covariates. The analysis was conducted in 2016. Of the total sample (n=253, survey response rate of 6.8%), the majority were non-Hispanic white (81.7%), married (74.5%), male (53.5%), and reported an annual household income of ≥ $50,000 (65.7%). Approximately, half were employed (49.6%), or had at least a college degree (51.8%). Slightly more than 1 in 5 (n=57, 22.5%) reported walking the recommended ≥150 minutes for any purpose in a typical week. The strongest predictors of walking the recommended ≥ 150 minutes for any purpose in a typical week in adjusted analysis were related to education and a high favorable perception of the neighborhood environment. Compared to those with a high school diploma or some college, participants with at least a college degree were five times as likely to walk the recommended ≥ 150 minutes for any purpose (OR=5.55, 95% CI=1.79-17.25). Walking the recommended ≥ 150 minutes for any purpose was significantly associated with participants who disagreed that there were many distracted drivers (e.g., on the cell phone while driving) in their neighborhood (OR=4.08, 95% CI=1.47-11.36) and those who agreed that there are sidewalks or protected walkways (e.g., walking trails) in their neighborhood (OR=3.55, 95% CI=1.10-11.49). Those employed were less likely to walk the recommended ≥ 150 minutes for any purpose compared to those unemployed (OR=0.31, 95% CI=0.11-0.85) as were those who reported some difficulty walking for a quarter of a mile (OR=0.19, 95% CI=0.05-0.77). Other socio-environmental factors such as having care-giver responsibilities for elders, someone to walk with, or a dog in the household as well as Walk Score™ were not significantly associated with walking the recommended ≥ 150 minutes for any purpose in a typical week. Neighborhood perception appears to be an important factor associated with the walking behaviors of middle-aged and older overweight and obese individuals. Enhancing the neighborhood environment (e.g., providing walking trails) may promote walking among these individuals.

Keywords: determinants of walking, obesity, older adults, physical activity

Procedia PDF Downloads 258

Authors: Nicolas R. Finkler, Wesley A. Saltarelli, Taison A. Bortolin, Vania E. Schneider, Davi G. F. Cunha

Abstract:

Estimating the relative contribution of nonpoint or point sources of pollution in low-orders streams is an important tool for the water resources management. The location of headwaters in areas with anthropogenic impacts from urbanization and agriculture is a common scenario in developing countries. This condition can lead to conflicts among different water users and compromise ecosystem services. Water pollution also contributes to exporting organic loads to downstream areas, including higher order rivers. The purpose of this research is to preliminarily assess nutrients and solids loads exported by water bodies located in watersheds with different types of land uses in São Carlos - SP (Latitude. -22.0087; Longitude. -47.8909) and Caxias do Sul - RS (Latitude. -29.1634, Longitude. -51.1796), Brazil, using regression analysis. The variables analyzed in this study were Total Kjeldahl Nitrogen (TKN), Nitrate (NO3-), Total Phosphorus (TP) and Total Suspended Solids (TSS). Data were obtained in October and December 2015 for São Carlos (SC) and in November 2012 and March 2013 for Caxias do Sul (CXS). Such periods had similar weather patterns regarding precipitation and temperature. Altogether, 11 sites were divided into two groups, some classified as more pristine (SC1, SC4, SC5, SC6 and CXS2), with predominance of native forest; and others considered as impacted (SC2, SC3, CXS1, CXS3, CXS4 and CXS5), presenting larger urban and/or agricultural areas. Previous linear regression was applied for data on flow and drainage area of each site (R² = 0.9741), suggesting that the loads to be assessed had a significant relationship with the drainage areas. Thereafter, regression analysis was conducted between the drainage areas and the total loads for the two land use groups. The R² values were 0.070, 0.830, 0.752 e 0.455 respectively for SST, TKN, NO3- and TP loads in the more preserved areas, suggesting that the loads generated by runoff are significant in these locations. However, the respective R² values for sites located in impacted areas were respectively 0.488, 0.054, 0.519 e 0.059 for SST, TKN, NO3- and P loads, indicating a less important relationship between total loads and runoff as compared to the previous scenario. This study suggests three possible conclusions that will be further explored in the full-text article, with more sampling sites and periods: a) In preserved areas, nonpoint sources of pollution are more significant in determining water quality in relation to the studied variables; b) The nutrient (TKN and P) loads in impacted areas may be associated with point sources such as domestic wastewater discharges with inadequate treatment levels; and c) The presence of NO3- in impacted areas can be associated to the runoff, particularly in agricultural areas, where the application of fertilizers is common at certain times of the year.

Keywords: land use, linear regression, point and non-point pollution sources, streams, water resources management

Procedia PDF Downloads 303

Authors: Kittipong Tissayakorn, Fumio Akagi, Yu Song

Abstract:

Logistics is the integrated management of all of the activities required to move products through the supply chain. For a typical product, this supply chain extends from a raw material source through the production and distribution system to the point of consumption and the associated reverse logistics. The logistical activities are comprised of freight transport, storage, inventory management, materials handling and all related information processing. This paper analyzes the green management system of logistics enterprise for Thailand and advances the concept of Green Logistics, which should be held by the public. In addition, it proposes that the government should strengthen its supervision and support for green logistics, and companies should construct self-disciplined green logistics management systems and corresponding processes, a reverse logistics management system and a modern green logistics information collection and management system.

Keywords: logistics, green logistics, management system, ecological economics

Procedia PDF Downloads 402

Authors: Azita Behbahaninia

Abstract:

This research has been carried out to determine the water quality and physico-chemical properties Polour River, one of the most branch of Haraz River. Polour River was studied for a period of one year Samples were taken from different stations along the main branch of River polour. In water samples determined pH, DO, SO4, Cl, PO4, NO3, EC, BOD, COD, Temprature, color and number of Caliform per liter. ArcGIS was used for the zoning of phosphate concentration in the polour River basin. The results indicated that the river is polluted in polour village station, because of discharge domestic wastewater and also river is polluted in Ziar village station, because of agricultural wastewater and water is contaminated in aquaculture station, because of fish ponds wastewater. Statistical analysis shows that between independent traits and coliform regression relationship is significant at the 1% level. Coefficient explanation index indicated independent traits control 80% coliform and 20 % is for unknown parameters. The causality analysis showed Temperature (0.6) has the most positive and direct effect on coliform and sulfate has direct and negative effect on coliform. The results of causality analysis and the results of the regression analysis are matched and other forms direct and indirect effects were negligible and ignorable. Kruskal-Wallis test showed, there is different between sampling stations and studied characters. Between stations for temperature, DO, COD, EC, sulfate and coliform is at 1 % and for phosphate 5 % level of significance.

Keywords: coliform, GIS, pollution, phosphate, river

Procedia PDF Downloads 466

Authors: Jaeyoung Lee, Minji Je

Abstract:

Objectives: In adolescence, adolescents are curious about sex, but sexual experience before becoming an adult can cause the risk of high probability of sexually transmitted infection. Therefore, it is very important to prevent sexually transmitted infections so that adolescents can grow in healthy and upright way. Adolescent females, especially, have sexual behavior distinguished from that of male adolescents. Protecting female adolescents’ reproductive health is even more important since it is directly related to the childbirth of the next generation. This study, thus, investigated the predictors of sexually transmitted infection in adolescent females with sexual experiences based on the National Health Statistics in Korea. Methods: This study was conducted based on the National Health Statistics in Korea. The 11th Korea Youth Behavior Web-based Survey in 2016 was conducted in the type of anonymous self-reported survey in order to find out the health behavior of adolescents. The target recruitment group was middle and high school students nationwide as of April 2016, and 65,528 students from a total of 800 middle and high schools participated. The study was conducted in 537 female high school students (Grades 10–12) among them. The collected data were analyzed as complex sampling design using SPSS statistics 22. The strata, cluster, weight, and finite population correction provided by Korea Center for Disease Control & Prevention (KCDC) were reflected to constitute complex sample design files, which were used in the statistical analysis. The analysis methods included Rao-Scott chi-square test, complex samples general linear model, and complex samples multiple logistic regression analysis. Results: Out of 537 female adolescents, 11.9% (53 adolescents) had experiences of venereal infection. The predictors for venereal infection of the subjects were ‘age at first intercourse’ and ‘sexual intercourse after drinking’. The sexually transmitted infection of the subjects was decreased by 0.31 times (p=.006, 95%CI=0.13-0.71) for middle school students and 0.13 times (p<.001, 95%CI=0.05-0.32) for high school students whereas the age of the first sexual experience was under elementary school age. In addition, the sexually transmitted infection of the subjects was 3.54 times (p < .001, 95%CI=1.76-7.14) increased when they have experience of sexual relation after drinking alcohol, compared to those without the experience of sexual relation after drinking alcohol. Conclusions: The female adolescents had high probability of sexually transmitted infection if their age for the first sexual experience was low. Therefore, the female adolescents who start sexual experience earlier shall have practical sex education appropriate for their developmental stage. In addition, since the sexually transmitted infection increases, if they have sexual relations after drinking alcohol, the consideration for prevention of alcohol use or intervention of sex education shall be required. When health education intervention is conducted for health promotion for female adolescents in the future, it is necessary to reflect the result of this study.

Keywords: adolescent, coitus, female, sexually transmitted diseases

Procedia PDF Downloads 191

Authors: Jamila Garba Audu, Shehu Usman Hassan

Abstract:

The widespread failure in the financial information quality has created the need to improve the financial information quality and to strengthen the control of managers by setting up good firms structures. Published accounting information in financial statements is required to provide various users - shareholders, employees, suppliers, creditors, financial analysts, stockbrokers and government agencies – with timely and reliable information useful for making prudent, effective and efficient decisions. The relationship between corporate governance and performance to financial reporting quality is imperative; this is because despite rapid researches in this area the findings obtained from these studies are constantly inconclusive. Data for the study were extracted from the firms’ annual reports and accounts. After running the OLS regression, a robustness test was conducted for the validity of statistical inferences; the data was empirically tested. A multiple regression was employed to test the model as a technique for data analysis. The results from the analysis revealed a negative association between all the regressors and financial reporting quality except the performance of listed manufacturing firms in Nigeria. This indicates that corporate governance plays a significant role in mitigating earnings management and improving financial reporting quality while performance does not. The study recommended among others that the composition of audit committee should be made in accordance with the provision for code of corporate governance which is not more than six (6) members with at least one (1) financial expert.

Keywords: corporate governance, financial reporting quality, manufacturing firms, Nigeria, performance

Procedia PDF Downloads 243

Authors: Golayeh Yousefi, Mehdi Homaee, Ali Akbar Norouzi

Abstract:

Ongoing monitoring of soil contamination by heavy metals is critical for ecosystem stability and environmental protection, and food security. The conventional methods of determining these soil contaminants are time-consuming and costly. Spectroscopy in the visible near-infrared (VNIR) - short wave infrared (SWIR) region is a rapid, non-destructive, noninvasive, and cost-effective method for assessment of soil heavy metals concentration by studying the spectral properties of soil constituents. The aim of this study is to derive spectral bands and important ranges that are sensitive to heavy metals and can be used to estimate the concentration of these soil contaminants. In other words, the change in the spectral properties of spectrally active constituents of soil can lead to the accurate identification and estimation of the concentration of these compounds in soil. For this purpose, 325 soil samples were collected, and their spectral reflectance curves were evaluated at a range of 350-2500 nm. After spectral preprocessing operations, the partial least-squares regression (PLSR) model was fitted on spectral data to predict the concentration of Cu and Ni. Based on the results, the spectral range of Cu- sensitive spectra were 480, 580-610, 1370, 1425, 1850, 1920, 2145, and 2200 nm, and Ni-sensitive ranges were 543, 655, 761, 1003, 1271, 1415, 1903, 2199 nm. Finally, the results of this study indicated that the spectral data contains a lot of information that can be applied to identify the soil properties, such as the concentration of heavy metals, with more detail.

Keywords: heavy metals, spectroscopy, spectral bands, PLS regression

Procedia PDF Downloads 82

Authors: Faiz N. K. Yusufi, Aquil Ahmed, Jamal Ahmad

Abstract:

Predicting the risk of diabetic retinopathy (DR) in Indian type 2 diabetes patients is immensely necessary. India, being the second largest country after China in terms of a number of diabetic patients, to the best of our knowledge not a single risk score for complications has ever been investigated. Diabetic retinopathy is a serious complication and is the topmost reason for visual impairment across countries. Any type or form of DR has been taken as the event of interest, be it mild, back, grade I, II, III, and IV DR. A sample was determined and randomly collected from the Rajiv Gandhi Centre for Diabetes and Endocrinology, J.N.M.C., A.M.U., Aligarh, India. Collected variables include patients data such as sex, age, height, weight, body mass index (BMI), blood sugar fasting (BSF), post prandial sugar (PP), glycosylated haemoglobin (HbA1c), diastolic blood pressure (DBP), systolic blood pressure (SBP), smoking, alcohol habits, total cholesterol (TC), triglycerides (TG), high density lipoprotein (HDL), low density lipoprotein (LDL), very low density lipoprotein (VLDL), physical activity, duration of diabetes, diet control, history of antihypertensive drug treatment, family history of diabetes, waist circumference, hip circumference, medications, central obesity and history of DR. Cox proportional hazard regression is used to design risk scores for the prediction of retinopathy. Model calibration and discrimination are assessed from Hosmer Lemeshow and area under receiver operating characteristic curve (ROC). Overfitting and underfitting of the model are checked by applying regularization techniques and best method is selected between ridge, lasso and elastic net regression. Optimal cut off point is chosen by Youden’s index. Five-year probability of DR is predicted by both survival function, and Markov chain two state model and the better technique is concluded. The risk scores developed can be applied by doctors and patients themselves for self evaluation. Furthermore, the five-year probabilities can be applied as well to forecast and maintain the condition of patients. This provides immense benefit in real application of DR prediction in T2DM.

Keywords: Cox proportional hazard regression, diabetic retinopathy, ROC curve, type 2 diabetes mellitus

Procedia PDF Downloads 183

Authors: Parsa Mahdavi, M. Amin Hariri-Ardebili

Abstract:

The development of accurate machine learning and deep learning models traditionally demands hands-on expertise and a solid background to fine-tune hyperparameters. With the continuous expansion of datasets in various scientific and engineering domains, researchers increasingly turn to machine learning methods to unveil hidden insights that may elude classic regression techniques. This surge in adoption raises concerns about the adequacy of the resultant meta-models and, consequently, the interpretation of the findings. In response to these challenges, automated machine learning (AutoML) emerges as a promising solution, aiming to construct machine learning models with minimal intervention or guidance from human experts. AutoML encompasses crucial stages such as data preparation, feature engineering, hyperparameter optimization, and neural architecture search. This paper provides a comprehensive overview of the principles underpinning AutoML, surveying several widely-used AutoML platforms. Additionally, the paper offers a glimpse into the application of AutoML on various engineering datasets. By comparing these results with those obtained through classical machine learning methods, the paper quantifies the uncertainties inherent in the application of a single ML model versus the holistic approach provided by AutoML. These examples showcase the efficacy of AutoML in extracting meaningful patterns and insights, emphasizing its potential to revolutionize the way we approach and analyze complex datasets.

Keywords: automated machine learning, uncertainty, engineering dataset, regression

Procedia PDF Downloads 60

Authors: Krishang Surapaneni

Abstract:

The goal of this project is to determine how to predict important aspects of options, including the ask price. We want to compare different machine learning models to learn the best model and the best hyperparameters for that model for this purpose and data set. Option pricing is a relatively new field, and it can be very complicated and intimidating, especially to inexperienced people, so we want to create a machine learning model that can predict important aspects of an option stock, which can aid in future research. We tested multiple different models and experimented with hyperparameter tuning, trying to find some of the best parameters for a machine-learning model. We tested three different models: a Random Forest Regressor, a linear regressor, and an MLP (multi-layer perceptron) regressor. The most important feature in this experiment is the ask price; this is what we were trying to predict. In the field of stock pricing prediction, there is a large potential for error, so we are unable to determine the accuracy of the models based on if they predict the pricing perfectly. Due to this factor, we determined the accuracy of the model by finding the average percentage difference between the predicted and actual values. We tested the accuracy of the machine learning models by comparing the actual results in the testing data and the predictions made by the models. The linear regression model performed worst, with an average percentage error of 17.46%. The MLP regressor had an average percentage error of 11.45%, and the random forest regressor had an average percentage error of 7.42%

Keywords: finance, linear regression model, machine learning model, neural network, stock price

Procedia PDF Downloads 74

Authors: Jacqueline Y. Thompson, Sam Watson, Lee Middleton, Karla Hemming

Abstract:

Introduction: The odds ratio (estimated via logistic regression) is a well-established and common approach for estimating covariate-adjusted binary treatment effects when comparing a treatment and control group with dichotomous outcomes. Its popularity is primarily because of its stability and robustness to model misspecification. However, the situation is different for the relative risk and risk difference, which are arguably easier to interpret and better suited to specific designs such as non-inferiority studies. So far, there is no equivalent, widely acceptable approach to estimate an adjusted relative risk and risk difference when conducting clinical trials. This is partly due to the lack of a comprehensive evaluation of available candidate methods. Methods/Approach: A simulation study is designed to evaluate the performance of relevant candidate methods to estimate relative risks to represent conditional and marginal estimation approaches. We consider the log-binomial, generalised linear models (GLM) with iteratively weighted least-squares (IWLS) and model-based standard errors (SE); log-binomial GLM with convex optimisation and model-based SEs; log-binomial GLM with convex optimisation and permutation tests; modified-Poisson GLM IWLS and robust SEs; log-binomial generalised estimation equations (GEE) and robust SEs; marginal standardisation and delta method SEs; and marginal standardisation and permutation test SEs. Independent and identically distributed datasets are simulated from a randomised controlled trial to evaluate these candidate methods. Simulations are replicated 10000 times for each scenario across all possible combinations of sample sizes (200, 1000, and 5000), outcomes (10%, 50%, and 80%), and covariates (ranging from -0.05 to 0.7) representing weak, moderate or strong relationships. Treatment effects (ranging from 0, -0.5, 1; on the log-scale) will consider null (H0) and alternative (H1) hypotheses to evaluate coverage and power in realistic scenarios. Performance measures (bias, mean square error (MSE), relative efficiency, and convergence rates) are evaluated across scenarios covering a range of sample sizes, event rates, covariate prognostic strength, and model misspecifications. Potential Results, Relevance & Impact: There are several methods for estimating unadjusted and adjusted relative risks. However, it is unclear which method(s) is the most efficient, preserves type-I error rate, is robust to model misspecification, or is the most powerful when adjusting for non-prognostic and prognostic covariates. GEE estimations may be biased when the outcome distributions are not from marginal binary data. Also, it seems that marginal standardisation and convex optimisation may perform better than GLM IWLS log-binomial.

Keywords: binary outcomes, statistical methods, clinical trials, simulation study

Procedia PDF Downloads 112

Authors: Saheed Shittu, Lolwa Alansari, Fahed Nattouf, Tawa Olukade, Naji Abdallah, Tamara Alshdafat, Sarra Amdouni

Abstract:

Cesarean sections (CS), a highly regarded surgical intervention for improving fetal-maternal outcomes and serving as an integral part of emergency obstetric services, are not without complications. Although CS has many advantages, it poses significant risks to both mother and child and increases healthcare expenditures in the long run. The escalating global prevalence of CS, coupled with variations in rates among immigrant populations, has prompted an inquiry into the correlation between CS rates and the nationalities of women undergoing deliveries at Al-Wakra Hospital (AWH), Qatar's second-largest public maternity hospital. This inquiry is motivated by the notable CS rate of 36%, deemed high in comparison to the 34% recorded across other Hamad Medical Corporation (HMC) maternity divisions This is Qatar's first comprehensive investigation of Caesarean section rates and nationalities. A retrospective cross-sectional study was conducted, and data for all births delivered in 2019 were retrieved from the hospital's electronic medical records. The CS rate, the crude rate, and adjusted risks of Caesarean delivery for mothers from each nationality were determined. The common indications for CS were analysed based on nationality. The association between nationality and Caesarean rates was examined using binomial logistic regression analysis considering Qatari women as a standard reference group. The correlation between the CS rate in the country of nationality and the observed CS rate in Qatar was also examined using Pearson's correlation. This study included 4,816 births from 69 different nationalities. CS was performed in 1767 women, equating to 36.5%. The nationalities with the highest CS rates were Egyptian (49.6%), Lebanese (45.5%), Filipino and Indian (both 42.2%). Qatari women recorded a CS rate of 33.4%. The major indication for elective CS was previous multiple CS (39.9%) and one prior CS, where the patient declined vaginal birth after the cesarean (VBAC) option (26.8%). A distinct pattern was noticed: elective CS was predominantly performed on Arab women, whereas emergency CS was common among women of Asian and Sub-Saharan African nationalities. Moreover, a significant correlation was found between the CS rates in Qatar and the women's countries of origin. Also, a high CS rate was linked to instances of previous CS. As a result of these insights, strategic interventions were successfully implemented at the facility to mitigate unwarranted CS, resulting in a notable reduction in CS rate from 36.5% in 2019 to 34% in 2022. This proves the efficacy of the meticulously researched approach. The focus has now shifted to reducing primary CS rates and facilitating well-informed decisions regarding childbirth methods.

Keywords: maternal nationality, caesarean section rate variation, migrants, high-income country

Procedia PDF Downloads 70

Authors: Siqi Fang, Man Cheung Chung

Abstract:

This study examined the inter-relationship between childhood maltreatment, cognitive distortion, alexithymia, eudemonic well-being, and psychological distress. One hundred and eighty-two university students participated in the study and completed an online survey comprising the Childhood Trauma Questionnaire, Cognitive Distortion Scale, Toronto Alexithymia Scale, Psychological Well-Being Scale, and General Health Questionnaire-28. Hierarchical multiple regression analysis showed that child maltreatment, perceptions of hopelessness and helplessness, preoccupation with danger, personal growth, and purpose in life predicted psychological distress. However, alexithymia was not a significant predictor. Further analysis using the regression models with bootstrapping procedure showed that feeling hopeless, helpless and preoccupation with danger mediated the path between child maltreatment and psychological distress. Meanwhile, coping with beliefs in personal growth and life purpose moderated the mediation effects of distorted cognition on psychological distress. To conclude, childhood maltreatment is associated with psychological distress. This relationship is influenced by people’s perceptions of life being hopeless, helpless or dangerous. At the same time, the effect of hopelessness, helplessness, and feelings of danger also depends on the degree of using coping strategies of positive psychological functioning.

Keywords: alexithymia, childhood maltreatment, cognitive distortion, eudemonic well-being, psychological distress

Procedia PDF Downloads 343

Authors: Artemis Louppis, Michalis Constantinou, Ioanna Kosma, Federica Blando, Michael Kontominas, Anastasia Badeka

Abstract:

The aim of the present study was to differentiate yellow prickly pear according to geographical origin based on the combination of mineral content, physicochemical parameters, vitamins and antioxidants. A total of 240 yellow prickly pear samples from Cyprus, Spain, Italy and Greece were analyzed for pH, titratable acidity, electrical conductivity, protein, moisture, ash, fat, antioxidant activity, individual antioxidants, sugars and vitamins by UPLC-MS/MS as well as minerals by ICP-MS. Statistical treatment of the data included multivariate analysis of variance followed by linear discriminant analysis. Based on results, a correct classification of 66.7% was achieved using the cross validation by mineral content while 86.1% was achieved using the cross validation method by combination of all analytical parameters.

Keywords: geographical differentiation, prickly pear, chemometrics, analytical techniques

Procedia PDF Downloads 141

Authors: Nancy Bartekian, Christine Reyna

Abstract:

Republicans and Democrats are often polarized when it comes to important topics, but the portrayal of polarization of key issues might be distorted and exaggerated. We examined Republicans' and Democrats’ attitudes about police reform policy during the 2020 racial justice protests and calls to ‘defund the police’. We hypothesized that a) Republicans and Democrats will be polarized on the “defund police'' question; however, b) they will have similar overall attitudes towards specific police reform policies (will be on the same side of the scale--disagree vs. agree), but c) will differ in their extent of agreement or disagreement (main effect of political party ID, but located on the same side of the scale). Using one-way, Multivariate analysis of covariance (MANCOVA) controlling for race, education, and income, we found an overall effect of political party ID. Six out of the nine policies studied were, in fact, not polarizing; both groups were in consensus on whether they disagreed or agreed with the policy, including “defund police''. Results suggest that polarization might be exaggerated.

Keywords: political psychology, social, ideology, polarization

Procedia PDF Downloads 100

Authors: Ivana Tadic, Snjezana Pivac

Abstract:

Although human resources are recognized as the crucial companies’ resources and their positive influence on companies’ performances has been confirmed through different researches, scientists are still debating it. In order to contribute this debate, this paper firstly discusses the most important human resource management elements and practices and its influence on companies’ success. Afterwards it defines human resource “bundles” – interrelated and internally consistent human resource practices, complementary to each other, or the most important human resource practices and elements regarding Croatian companies and its human resource management activities. Finally, the paper provides empirical results; more precisely it reveals the relation of the level of development of human resource management function (“bundles”) and companies’ financial performances (using profitability ratios, liquidity ratios, solvency ratios and a group of additional ratios related to employees’ indicators).

Keywords: companies’ performances, human resource bundles, multivariate statistical analysis, marketing

Procedia PDF Downloads 421

Authors: Mahshid Rezaei, Zahra Tajari, Zahra Esmaeily, Atefeh Eyvazkhani, Shahrzad Daei, Marjan Mansouri Dara, Mohaddesh Rezaei, Abolghassem Djazayeri, Ahmadreza Dorosti Motlagh

Abstract:

Background: Musculoskeletal pain is one of the most prevalent symptoms in elderly age. Nutrition and diet are considered important underlying factors that could affect chronic musculoskeletal pain. The purpose of this study was to identify the relationship between empirical dietary inflammatory patterns (EDII) and musculoskeletal pain. Method: In this cross-sectional study, 213 elderly individuals were selected from several health centers. The usual dietary intake was evaluated by a valid and reliable 147-items food frequency questionnaire (FFQ). To measure the intensity of pain, Visual Analogue Scale (VAS) was used. Multiple Linear Regression was applied to assess the association between EDII and musculoskeletal pain. Results: The results of multiple linear regression analysis indicate that a higher EDII score was associated with higher musculoskeletal pain (β= 0.21: 95% CI: 0.24-1.87: P= 0.003). These results stayed significant even after adjusting for covariates such as sex, marital status, height, family number, sleep, BMI, physical activity duration, waist circumference, protector, and medication use (β= 0.16: 95% CI: 0.11-1.04: P= 0.02). Conclusion: Study findings indicated that higher inflammation of diet might have a direct association with musculoskeletal pains in elderlies. However, further investigations are required to confirm these findings.

Keywords: musculoskeletal pain, empirical dietary inflammatory pattern, elderlies, dietary pattern

Procedia PDF Downloads 208

Authors: Mercedes Bertotto, Marcelo Bello, Hector Goicoechea, Veronica Fusca

Abstract:

The National Service of Agri-Food Health and Quality (SENASA), controls honey to detect contamination by synthetic or natural chemical substances and establishes and controls the traceability of the product. The utility of near-infrared spectroscopy for the detection of adulteration of honey with high fructose corn syrup (HFCS) was investigated. First of all, a mixture of different authentic artisanal Argentinian honey was prepared to cover as much heterogeneity as possible. Then, mixtures were prepared by adding different concentrations of high fructose corn syrup (HFCS) to samples of the honey pool. 237 samples were used, 108 of them were authentic honey and 129 samples corresponded to honey adulterated with HFCS between 1 and 10%. They were stored unrefrigerated from time of production until scanning and were not filtered after receipt in the laboratory. Immediately prior to spectral collection, honey was incubated at 40°C overnight to dissolve any crystalline material, manually stirred to achieve homogeneity and adjusted to a standard solids content (70° Brix) with distilled water. Adulterant solutions were also adjusted to 70° Brix. Samples were measured by NIR spectroscopy in the range of 650 to 7000 cm⁻¹. The technique of specular reflectance was used, with a lens aperture range of 150 mm. Pretreatment of the spectra was performed by Standard Normal Variate (SNV). The ant colony optimization genetic algorithm sample selection (ACOGASS) graphical interface was used, using MATLAB version 5.3, to select the variables with the greatest discriminating power. The data set was divided into a validation set and a calibration set, using the Kennard-Stone (KS) algorithm. A combined method of Potential Functions (PF) was chosen together with Partial Least Square Linear Discriminant Analysis (PLS-DA). Different estimators of the predictive capacity of the model were compared, which were obtained using a decreasing number of groups, which implies more demanding validation conditions. The optimal number of latent variables was selected as the number associated with the minimum error and the smallest number of unassigned samples. Once the optimal number of latent variables was defined, we proceeded to apply the model to the training samples. With the calibrated model for the training samples, we proceeded to study the validation samples. The calibrated model that combines the potential function methods and PLSDA can be considered reliable and stable since its performance in future samples is expected to be comparable to that achieved for the training samples. By use of Potential Functions (PF) and Partial Least Square Linear Discriminant Analysis (PLS-DA) classification, authentic honey and honey adulterated with HFCS could be identified with a correct classification rate of 97.9%. The results showed that NIR in combination with the PT and PLS-DS methods can be a simple, fast and low-cost technique for the detection of HFCS in honey with high sensitivity and power of discrimination.

Keywords: adulteration, multivariate analysis, potential functions, regression

Procedia PDF Downloads 124

Authors: Elinor Bull, Faye Horsley

Abstract:

Abstract Objective: The current study investigated if there was a difference in personality factors in prisoners who had a recorded history of firesetting and who had no recorded history of firesetting. Participants: Participants were 64 male prisoners in a Category B male prison. Participants who had set a fire were identified through the prisons data base, and prisoners who had not set a fire were selected at random. Method: The study used the International Personality Item Pool-50 to measure personality factors, and prisoners who had set a fire were identified through a range of sources accessible to the prison. Analytical evaluation was done by the Multivariate Kruskal Wallis and Mann-Whitney tests. Findings: There was a significant difference between the the firesetting and non-firesetting group in the scores of the personality factor of Contentiousness. Contentiousness was significantly lower in the firesetting sample compared to the non-firesetting sample. Conclusions: Implications for clinical practice and future research are discussed.

Keywords: firesetting, personality, arson, prison, prisoners

Procedia PDF Downloads 82

Authors: Nguyen Ngoc Que Tran

Abstract:

Purpose: The purpose of this paper is to investigate the relationship between internal market orientation, external market orientation, employee commitment and job satisfaction. Design/methodology/approach: This study collected data through a survey and utilized simple linear regression and multiple regression analysis to determine if there was any support for the research hypotheses as presented in the previous chapter. Findings: Using data from 256 employees of four leading joint stock banks in Vietnam, the empirical results indicates that employee commitment is positively related with external market orientation, job satisfaction is positively related to employee commitment, and employee commitment and job satisfaction are positively related to external market orientation. However, job satisfaction has no significant positive effect on external market orientation. Theoretical contribution: The primary contribution to marketing theory arising from this study is the integration of job satisfaction, employee commitment, and external market orientation in a single research model. Practical implications: The major contribution to practice is an external market oriented bank has to respond rapidly to the future needs and preferences of its customers. This could result in high levels of commitment to the service process and in doing so provide Vietnamese joint-stock commercial banks with a competitive advantage. The finding is important for the banking service sector in general and the Vietnamese banking industry in particular.

Keywords: employee commitment, job satisfaction and external market orientation, vietnam, bank

Procedia PDF Downloads 412

Authors: Petr Gurný

Abstract:

One of the most important tasks in the risk management is the correct determination of probability of default (PD) of particular financial subjects. In this paper a possibility of determination of financial institution’s PD according to the credit-scoring models is discussed. The paper is divided into the two parts. The first part is devoted to the estimation of the three different models (based on the linear discriminant analysis, logit regression and probit regression) from the sample of almost three hundred US commercial banks. Afterwards these models are compared and verified on the control sample with the view to choose the best one. The second part of the paper is aimed at the application of the chosen model on the portfolio of three key Czech banks to estimate their present financial stability. However, it is not less important to be able to estimate the evolution of PD in the future. For this reason, the second task in this paper is to estimate the probability distribution of the future PD for the Czech banks. So, there are sampled randomly the values of particular indicators and estimated the PDs’ distribution, while it’s assumed that the indicators are distributed according to the multidimensional subordinated Lévy model (Variance Gamma model and Normal Inverse Gaussian model, particularly). Although the obtained results show that all banks are relatively healthy, there is still high chance that “a financial crisis” will occur, at least in terms of probability. This is indicated by estimation of the various quantiles in the estimated distributions. Finally, it should be noted that the applicability of the estimated model (with respect to the used data) is limited to the recessionary phase of the financial market.

Keywords: credit-scoring models, multidimensional subordinated Lévy model, probability of default

Procedia PDF Downloads 452

Authors: Anuradha Chug, Sunali Gandhi

Abstract:

Regression Testing is a very expensive and time-consuming process carried out to ensure the validity of modified software. Due to the availability of insufficient resources to re-execute all the test cases in time constrained environment, efforts are going on to generate test data automatically without human efforts. Many search based techniques have been proposed to generate efficient, effective as well as optimized test data, so that the overall cost of the software testing can be minimized. The generated test data should be able to uncover all potential lapses that exist in the software or product. Inspired from the natural behavior of bat for searching her food sources, current study employed a meta-heuristic, search-based bat algorithm for optimizing the test data on the basis certain parameters without compromising their effectiveness. Mathematical functions are also applied that can effectively filter out the redundant test data. As many as 50 Java programs are used to check the effectiveness of proposed test data generation and it has been found that 86% saving in testing efforts can be achieved using bat algorithm while covering 100% of the software code for testing. Bat algorithm was found to be more efficient in terms of simplicity and flexibility when the results were compared with another nature inspired algorithms such as Firefly Algorithm (FA), Hill Climbing Algorithm (HC) and Ant Colony Optimization (ACO). The output of this study would be useful to testers as they can achieve 100% path coverage for testing with minimum number of test cases.

Keywords: regression testing, test case selection, test case prioritization, genetic algorithm, bat algorithm

Procedia PDF Downloads 379

Authors: Mehdi Mirzaei Alavijeh, Farzad Jalilian

Abstract:

Background: Marijuana is the most widely used illicit drug worldwide, especially among adolescents and young adults, which can cause numerous complications. The aim of this study was to determine the pattern, motivation use, and factors related to marijuana use among Iranian youths based on the integrative model of behavioral prediction Methods: A cross-sectional study was conducted among 174 youths marijuana user in Kermanshah County and Isfahan County, during summer 2014 which was selected with the convenience sampling for participation in this study. A self-reporting questionnaire was applied for collecting data. Data were analyzed by SPSS version 21 using bivariate correlations and linear regression statistical tests. Results: The mean marijuana use of respondents was 4.60 times at during week [95% CI: 4.06, 5.15]. Linear regression statistical showed, the structures of integrative model of behavioral prediction accounted for 36% of the variation in the outcome measure of the marijuana use at during week (R2 = 36% & P < 0.001); and among them attitude, marijuana refuse, and subjective norms were a stronger predictors. Conclusion: Comprehensive health education and prevention programs need to emphasize on cognitive factors that predict youth’s health-related behaviors. Based on our findings it seems, designing educational and behavioral intervention for reducing positive belief about marijuana, marijuana self-efficacy refuse promotion and reduce subjective norms encourage marijuana use has an effective potential to protect youths marijuana use.

Keywords: marijuana, youth, integrative model of behavioral prediction, Iran

Procedia PDF Downloads 553