Abstract The use of credit for various occasions has become a routine in our lives. In return, banking and financial institutions require to determine whether the loan demands from them contain any risk. Accordingly, these institutions have been increased their activities in determining whether credit rating models from past credit records of the person applying for the loan works properly. Machine learning-based technologies have opened a new era in this field. AI and machine learning based methods for credit scoring are currently implemented by banking or non-banking financial institutions. Employed models are to extract meaningful features from the required data in which wide variety of information available. In this study, credit risk assessment is conducted using boosting methods such as CatBoost, XGBoost and Light GBM. To this aim, Kaggle Home Credit Default Risk dataset is used and the effect of crediting tendency on the results is also considered. The results have shown that gradient boosting methods provide results that are close to each other, and crediting tendency produces better AUC score in CatBoost while it causes a small decrement in AUC score of XGBoost and LightGBM.
{"title":"Credit risk analysis using boosting methods","authors":"S. Coşkun, M. Turanli","doi":"10.2478/jamsi-2023-0001","DOIUrl":"https://doi.org/10.2478/jamsi-2023-0001","url":null,"abstract":"Abstract The use of credit for various occasions has become a routine in our lives. In return, banking and financial institutions require to determine whether the loan demands from them contain any risk. Accordingly, these institutions have been increased their activities in determining whether credit rating models from past credit records of the person applying for the loan works properly. Machine learning-based technologies have opened a new era in this field. AI and machine learning based methods for credit scoring are currently implemented by banking or non-banking financial institutions. Employed models are to extract meaningful features from the required data in which wide variety of information available. In this study, credit risk assessment is conducted using boosting methods such as CatBoost, XGBoost and Light GBM. To this aim, Kaggle Home Credit Default Risk dataset is used and the effect of crediting tendency on the results is also considered. The results have shown that gradient boosting methods provide results that are close to each other, and crediting tendency produces better AUC score in CatBoost while it causes a small decrement in AUC score of XGBoost and LightGBM.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"19 1","pages":"5 - 18"},"PeriodicalIF":0.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45360964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract In this study, in order to prevent information loss, we propose two dissimilarity measures between intuitionistic fuzzy sets (IFSs), which consider membership and non-membership degree and IFSs is farther extension of Fuzzy sets (FSs). Additionally, we have examined the characteristics of the proposed metrics to confirm their validity. We then conducted a series of experiments, including numerical experimentation, pattern recognition, and clustering analysis, to evaluate the efficacy of these metrics. The comparative outcomes illustrate that our dissimilarity metrics are more straightforward, easy to understand, and superior to the majority of the existing methods.
{"title":"Dissimilarity measure between intuitionistic Fuzzy sets and its applications in pattern recognition and clustering analysis","authors":"V. Rani, S. Kumar","doi":"10.2478/jamsi-2023-0004","DOIUrl":"https://doi.org/10.2478/jamsi-2023-0004","url":null,"abstract":"Abstract In this study, in order to prevent information loss, we propose two dissimilarity measures between intuitionistic fuzzy sets (IFSs), which consider membership and non-membership degree and IFSs is farther extension of Fuzzy sets (FSs). Additionally, we have examined the characteristics of the proposed metrics to confirm their validity. We then conducted a series of experiments, including numerical experimentation, pattern recognition, and clustering analysis, to evaluate the efficacy of these metrics. The comparative outcomes illustrate that our dissimilarity metrics are more straightforward, easy to understand, and superior to the majority of the existing methods.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"19 1","pages":"61 - 77"},"PeriodicalIF":0.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48130630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract In this paper, we look at the general form of the two-point quadrature formulas, which doesn’t have to be symmetric. Under the convexity constraint of the first derivative, we propose a new scheme that gives the best approximation of the two-point quadrature rules.
{"title":"Refinement of the general form of the two-point quadrature formulas via convexity","authors":"D. C. Benchettah, A. Lakhdari, B. Meftah","doi":"10.2478/jamsi-2023-0006","DOIUrl":"https://doi.org/10.2478/jamsi-2023-0006","url":null,"abstract":"Abstract In this paper, we look at the general form of the two-point quadrature formulas, which doesn’t have to be symmetric. Under the convexity constraint of the first derivative, we propose a new scheme that gives the best approximation of the two-point quadrature rules.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"19 1","pages":"93 - 101"},"PeriodicalIF":0.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42641390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract In this paper, based on a given parameterized identity that generates a quadrature rule family similar to Simpson’s second formula, we establish some new Simpson-like type inequalities for functions with bounded as well as Lipchitzian derivatives from which we can deduce the famous 3/8-Simpson’s inequality. The study concludes with an application example from management science.
{"title":"Parameterized Simpson-like inequalities for differentiable Bounded and Lipschitzian functions with application example from management science","authors":"N. Boutelhig, B. Meftah, W. Saleh, A. Lakhdari","doi":"10.2478/jamsi-2023-0005","DOIUrl":"https://doi.org/10.2478/jamsi-2023-0005","url":null,"abstract":"Abstract In this paper, based on a given parameterized identity that generates a quadrature rule family similar to Simpson’s second formula, we establish some new Simpson-like type inequalities for functions with bounded as well as Lipchitzian derivatives from which we can deduce the famous 3/8-Simpson’s inequality. The study concludes with an application example from management science.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"19 1","pages":"79 - 91"},"PeriodicalIF":0.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47601111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract A new Lomax Rayleigh distribution (NLRD) is proposed and generated using Transformed Transformer (T-X) family generator. Various structural properties like generating functions, moments, limiting form, quantile function median and mode of NLRD are studied. Maximum likelihood estimators (MLEs) of the parameters are obtained and the model fitting is tested with simulated data. Model adequacy with live data is explained with two real-time cancer data sets. The NLRD shows a better fit in the estimation of survival in bile duct cancer and head and neck cancer data than other existing distributions.
{"title":"Survival analysis of cancer patients using a new Lomax Rayleigh distribution","authors":"K. Naga Saritha, G. S. Rao, K. Rosaiah","doi":"10.2478/jamsi-2023-0002","DOIUrl":"https://doi.org/10.2478/jamsi-2023-0002","url":null,"abstract":"Abstract A new Lomax Rayleigh distribution (NLRD) is proposed and generated using Transformed Transformer (T-X) family generator. Various structural properties like generating functions, moments, limiting form, quantile function median and mode of NLRD are studied. Maximum likelihood estimators (MLEs) of the parameters are obtained and the model fitting is tested with simulated data. Model adequacy with live data is explained with two real-time cancer data sets. The NLRD shows a better fit in the estimation of survival in bile duct cancer and head and neck cancer data than other existing distributions.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"19 1","pages":"19 - 45"},"PeriodicalIF":0.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44045217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Kyselica, J. Silha, R. Ďurikovič, D. Bartková, J. Tóth
Abstract The space debris population is constantly increasing, including large spare upper stages and spacecraft. Atmospheric re-entry events caused by these artificial objects pose a threat to the ground-based population and their understanding and accurate modelling are required. On the 24th of October 2020, two cameras of the AMOS system in the Hawaiian Islands recorded the re-entry event of spare rocket’s upper stage. In the presented work, we aim to improve and automate the process of fragment identification between simultaneously acquired video recordings of this event. We are presenting the whole processing pipeline. First segmentation is performed to identify the cluster of interest and flying fragments within that cluster in all frames. The acquired information is used to create tracks of individual flying fragments – frame objects to be further associated between recordings.
{"title":"Towards image processing of reentry event","authors":"D. Kyselica, J. Silha, R. Ďurikovič, D. Bartková, J. Tóth","doi":"10.2478/jamsi-2023-0003","DOIUrl":"https://doi.org/10.2478/jamsi-2023-0003","url":null,"abstract":"Abstract The space debris population is constantly increasing, including large spare upper stages and spacecraft. Atmospheric re-entry events caused by these artificial objects pose a threat to the ground-based population and their understanding and accurate modelling are required. On the 24th of October 2020, two cameras of the AMOS system in the Hawaiian Islands recorded the re-entry event of spare rocket’s upper stage. In the presented work, we aim to improve and automate the process of fragment identification between simultaneously acquired video recordings of this event. We are presenting the whole processing pipeline. First segmentation is performed to identify the cluster of interest and flying fragments within that cluster in all frames. The acquired information is used to create tracks of individual flying fragments – frame objects to be further associated between recordings.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"19 1","pages":"47 - 60"},"PeriodicalIF":0.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41661228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Climate change is a significant contributor to environmental harm and the rise in Atmospheric carbon dioxide, which raises the earth’s surface temperature. As forests are the primary mechanism for absorbing carbon dioxide gas and protecting the earth from global warming and unpredictable weather patterns, a high rate of deforestation is to blame for this. In this study, the economic drivers causing deforestation in Tanzania include per capita income, per capita purchasing power, inflation rate, per capita purchasing power, poverty rate, and electricity consumption are investigated. Autoregressive models with exogenous variables (VARX (1) – VARX (3)) models are formulated to analyze the effect of economic variables and forecast the rate of deforestation in Tanzania. The time series data used from 1994 to 2014 were collected in Tanzania, nature of the data suggests the increase in the rate of deforestation as time progresses. In this study, the best model VARX (3, 0) was obtained, and the relationship between the variables through granger causality was obtained. Also, forecasting was carried out for the next 10 years using the best model VARX (3, 0) and Kalman Filters. It was observed that economic variables, especially the poverty rate, have an impact on the rate of deforestation in Tanzania. Furthermore, the graph shows the increasing trend in the rate of deforestation in the coming years in Tanzania.
{"title":"Time Series Analysis of Economic Factors Influencing Deforestation in Tanzania","authors":"John Gweba, I. Mbalawata, S. Mirau","doi":"10.22457/jmi.v24a06218","DOIUrl":"https://doi.org/10.22457/jmi.v24a06218","url":null,"abstract":"Climate change is a significant contributor to environmental harm and the rise in Atmospheric carbon dioxide, which raises the earth’s surface temperature. As forests are the primary mechanism for absorbing carbon dioxide gas and protecting the earth from global warming and unpredictable weather patterns, a high rate of deforestation is to blame for this. In this study, the economic drivers causing deforestation in Tanzania include per capita income, per capita purchasing power, inflation rate, per capita purchasing power, poverty rate, and electricity consumption are investigated. Autoregressive models with exogenous variables (VARX (1) – VARX (3)) models are formulated to analyze the effect of economic variables and forecast the rate of deforestation in Tanzania. The time series data used from 1994 to 2014 were collected in Tanzania, nature of the data suggests the increase in the rate of deforestation as time progresses. In this study, the best model VARX (3, 0) was obtained, and the relationship between the variables through granger causality was obtained. Also, forecasting was carried out for the next 10 years using the best model VARX (3, 0) and Kalman Filters. It was observed that economic variables, especially the poverty rate, have an impact on the rate of deforestation in Tanzania. Furthermore, the graph shows the increasing trend in the rate of deforestation in the coming years in Tanzania.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"9 1","pages":""},"PeriodicalIF":0.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80736331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the field of computer-aided design and related applications, the free curve with parameters shows its strong function. But an excellent curve with parameters need to eliminate unnecessary cusps and inflection points, so shape analysis is needed for some specific curves. We have already shown many excellent properties of the DTB-like curves and DT B-spline-like curves. Thus the shape features like convexity, loops, inflection points, and cusps of the DTB-like curves, and DT B-spline-like curves are further discussed in this paper. And the necessary and sufficient conditions for the existence of shape features of the corresponding DTB-like curves and DT B-spline-like curves are given. And the shape features distribution all be generalized into tables and diagrams, which are useful for industrial design. In addition, the effect of shape parameters on the shape features diagram and its adjustment ability to the shape of corresponding curves are analyzed, respectively. The work in this paper enables users to determine how to set parameter values so that the generated curve could be a global convex or local convex curve, has a required inflection point, or eliminate the unnecessary one, and could be adjusted to another aimed shape.
{"title":"The Shape Analysis of DTB-like and DT B-spline-like Curves","authors":"Zhang Hangjian, Zhang Guicang, Wang Lu","doi":"10.22457/jmi.v24a01216","DOIUrl":"https://doi.org/10.22457/jmi.v24a01216","url":null,"abstract":"In the field of computer-aided design and related applications, the free curve with parameters shows its strong function. But an excellent curve with parameters need to eliminate unnecessary cusps and inflection points, so shape analysis is needed for some specific curves. We have already shown many excellent properties of the DTB-like curves and DT B-spline-like curves. Thus the shape features like convexity, loops, inflection points, and cusps of the DTB-like curves, and DT B-spline-like curves are further discussed in this paper. And the necessary and sufficient conditions for the existence of shape features of the corresponding DTB-like curves and DT B-spline-like curves are given. And the shape features distribution all be generalized into tables and diagrams, which are useful for industrial design. In addition, the effect of shape parameters on the shape features diagram and its adjustment ability to the shape of corresponding curves are analyzed, respectively. The work in this paper enables users to determine how to set parameter values so that the generated curve could be a global convex or local convex curve, has a required inflection point, or eliminate the unnecessary one, and could be adjusted to another aimed shape.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"45 1","pages":""},"PeriodicalIF":0.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76385866","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we concern with the blow-up problem of positive solutions to parabolic equations with reaction terms of local and nonlocal type involving a variable exponent. It is shown that under certain conditions on the nonlinearities and data, blowups will occur for the limited time.
{"title":"Blow-up for Semilinear Parabolic Equations with Memory Terms of Variable Exponents","authors":"Xiang Jing, Tian Ya","doi":"10.22457/jmi.v24a04217","DOIUrl":"https://doi.org/10.22457/jmi.v24a04217","url":null,"abstract":"In this paper, we concern with the blow-up problem of positive solutions to parabolic equations with reaction terms of local and nonlocal type involving a variable exponent. It is shown that under certain conditions on the nonlinearities and data, blowups will occur for the limited time.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"60 1","pages":""},"PeriodicalIF":0.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76022121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Asymmetric information propagation frequently results in distorted financial markets and is generally a feature of informationally inefficient markets. We developed a model of optimal asset allocation using the martingale method to assist an investor in selecting an asset that performs better under the conditions of a market information cascade. In order to confirm that the model satisfies the required conditions, we applied the verification theorem and ascertained that the results produced were optimal. The model outperformed the famous Markowitz mean-variance type model and was shown to produce stable and consistent solutions under such market conditions.
{"title":"Portfolio Optimization under Informationally Asymmetric Markets","authors":"Abdulmaliki Aruna Kilange, Kube Ananda, P. Ngare","doi":"10.22457/jmi.v24a05220","DOIUrl":"https://doi.org/10.22457/jmi.v24a05220","url":null,"abstract":"Asymmetric information propagation frequently results in distorted financial markets and is generally a feature of informationally inefficient markets. We developed a model of optimal asset allocation using the martingale method to assist an investor in selecting an asset that performs better under the conditions of a market information cascade. In order to confirm that the model satisfies the required conditions, we applied the verification theorem and ascertained that the results produced were optimal. The model outperformed the famous Markowitz mean-variance type model and was shown to produce stable and consistent solutions under such market conditions.","PeriodicalId":43016,"journal":{"name":"Journal of Applied Mathematics Statistics and Informatics","volume":"68 1","pages":""},"PeriodicalIF":0.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87510680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}