5 results on '"RANDOM forest algorithms"'
Search Results
2. An Approach for Variable Selection and Prediction Model for Estimating the Risk-Based Capital (RBC) Based on Machine Learning Algorithms
- Author
-
Jaewon Park and Minsoo Shin
- Subjects
life insurance companies ,Bayesian Regulatory Neural Network ,corporate sustainable management ,Random Forest algorithms ,Strategy and Management ,Economics, Econometrics and Finance (miscellaneous) ,RBC ratio ,Machine learning ,Insurance ,Accounting ,HG8011-9999 ,ddc:330 - Abstract
The risk-based capital (RBC) ratio, an insurance company’s financial soundness system, evaluates the capital adequacy needed to withstand unexpected losses. Therefore, continuous institutional improvement has been made to monitor the financial solvency of companies and protect consumers’ rights, and improvement of solvency systems has been researched. The primary purpose of this study is to find a set of important predictors to estimate the RBC ratio of life insurance companies in a large number of variables (1891), which includes crucial finance and management indices collected from all Korean insurers quarterly under regulation for transparent management information. This study employs a combination of Machine learning techniques: Random Forest algorithms and the Bayesian Regulatory Neural Network (BRNN). The combination of Random Forest algorithms and BRNN predicts the next period’s RBC ratio better than the conventional statistical method, which uses ordinary least-squares regression (OLS). As a result of the findings from Machine learning techniques, a set of important predictors is found within three categories: liabilities and expenses, other financial predictors, and predictors from business performance. The dataset of 23 companies with 1891 variables was used in this study from March 2008 to December 2018 with quarterly updates for each year.
- Published
- 2022
- Full Text
- View/download PDF
3. Master DISD - Machine Learning e qualità dell'aria Analisi ed addestramento di algoritmi per la previsione del PM10
- Author
-
Alessandro Lotti and Giampalo Liuzzi
- Subjects
machine learning ,random forest algorithms ,data intelligence ,air quality ,environment - Abstract
Il monitoraggio degli inquinanti atmosferici, quali PM10 o il monossido di azoto, è uno degli aspetti chiave nella tutela dell’ambiente che da diversi anni è al centro di dibattiti e politiche poste in campo da amministrazioni locali e nazionali, finalizzate alla lotta alle emissioni. L’inquinamento dell’aria in particolare è dato dalla contaminazione da parte di agenti chimici, fisici e/o biologici che modificano le caratteristiche naturali dell’atmosfera. Gli apparecchi per il riscaldamento delle abitazioni, i motori dei veicoli, gli impianti industriali, sono comuni sorgenti di inquinamento atmosferico. In Italia, le emissioni sono diminuite notevolmente negli ultimi decenni, con conseguente miglioramento della qualità dell'aria; tuttavia, le concentrazioni di inquinanti atmosferici sono ancora troppo elevate e i problemi di qualità dell'aria persistono soprattutto nelle grandi città. Il rapporto tra emissioni e concentrazioni in atmosfera non è diretto e lineare ma la sua variabilità nel tempo e nello spazio dipendono infatti, oltre che dal carico emissivo, da altri fattori, legati alla meteorologia e alla reattività chimica delle sostanze emesse. Nel corso degli anni, l’analisi delle serie storiche delle emissioni e degli inventari nazionali ha permesso quindi di capire meglio i fenomeni e mettere in campo una serie di misure, azioni, piani e programmi di risanamento della qualità dell'aria. Inoltre, negli ultimi anni lo sviluppo della “scienza dei dati” ed in particolare degli algoritmi di apprendimento automatico hanno permesso di esplorare nuovi campi nello studio di questi fenomeni e di supportare le amministrazioni locali nella pianificazione delle suddette azioni di risanamento e mitigazione. Il seguente lavoro è la tesi finale del Master di II Livello Data Intelligence e Strategie Decisionali presso l'Università degli Studi di Roma Sapienza. Si è cercato di approfondire le possibili applicazioni di questo nuovo approccio al caso studio della città di Roma ed utilizzando due tra i più comuni algoritmi, Random Forest e Decision tree, costruendo un modello in grado di prevedere con una certa accuratezza i valori del PM10. I risultati sono stati poi riportati su una dashboard di tipo spaziale al fine di creare uno strumento di supporto alle decisioni per le amministrazioni comunali e di pianificazione di politiche di riduzione e contenimento dell’inquinamento atmosferico., {"references":["Biau G., Devroye L. and Lugosi G. \"Consistency of Random Forests and Other Averaging Classifiers\" Journal of Machine Learning Research 9 (2008) 2015-2033","Breiman, L. Random Forests. Machine Learning 45, 5–32 (2001)","Decreto Legislativo 13 agosto 2010, n.155 \"Attuazione della direttiva 2008/50/CE relativa alla qualità dell'aria ambiente e per un'aria più pulita in Europa\"","Grange, S. K., Carslaw, D. C., Lewis, A. C., Boleti, E., and Hueglin, C. \"Random forest meteorological normalisation models for Swiss PM10 trend analysis\", Atmos. Chem. Phys., 18, 6223–6239","L. Grippo and M. Sciandrone \"Metodi di ottimizzazione per le reti neurali\"","Lg. 28 giugno 2016 n. 132 – Istituzione del Sistema nazionale a rete per la protezione dell'ambiente e disciplina dell'Istituto Superiore per la Protezione e la Ricerca Ambientale","Kotsiantis, S.B. \"Decision trees: a recent overview\" Artif Intell Rev 39, 261–283 (2013)","Navares R., Aznarte L. J. \"Predicting air quality with deep learning LSTM: Towards comprehensive models\" Ecological Informatics 55 (2020)","Zachary M. Jones and Fridolin J. Linder \"Exploratory Data Analysis using Random Forests\" The journal of open source software","ISPRA – SNPA \"Ambiente in Italia – Trend e Normative\"","Report NPA n. 17/2020 \"La qualità dell'aria in Italia. Edizione 2020\""]}
- Published
- 2021
- Full Text
- View/download PDF
4. Machine Learning: An overview with the help of R software
- Author
-
IJSMI, Editor
- Subjects
K-Nearest Neighborhood Algorithms ,Decision Tree Algorithms ,Clustering Algorithms ,machine learning ,Regression Algorithms ,Artificial Neural Network Algorithms ,Naïve Bayes Algorithms ,Random Forest Algorithms ,GeneralLiterature_REFERENCE(e.g.,dictionaries,encyclopedias,glossaries) ,Support Vector Machine Algorithms - Abstract
Machine Learning: An overview with the help of R software Preface This book intends to provide an overview of Machine Learning and its algorithms & models with help of R software. Machine learning forms the basis for Artificial Intelligence which will play a crucial role in day to day life of human beings in the near future. A basic understanding of machine learning is required, as its application is widely seen in different fields such as banks and financial sectors, manufacturing, aviation, transportation and medical field. The book covers machine learning classification algorithms such as K-Nearest Neighborhood, Naïve Bayes, Decision Trees and also Artificial Neural Networks and Support Vector Machines. It is recommended to refer author’s book on Application of Statistical Tools in Biomedical Domain: An Overview with Help of Software (https://www.amazon.com/dp/1986988554) and Essentials of Bio-Statistics: An overview with the help of Software https://www.amazon.com/dp/B07GRBXX7D if you need to familiarize yourself with the basic statistical knowledge. Editor International Journal of Statistics and Medical Informatics www.ijsmi.com/book.php Amazon link https://www.amazon.com/dp/1790122627 (Paper Back) https://www.amazon.com/dp/B07KQSN447 (Kindle Edition)  
- Published
- 2018
- Full Text
- View/download PDF
5. Estimating the BIS Capital Adequacy Ratio for Korean Banks Using Machine Learning: Predicting by Variable Selection Using Random Forest Algorithms
- Author
-
Wookjae Heo, Jaewon Park, and Minsoo Shin
- Subjects
Index (economics) ,Computer science ,bank ,Strategy and Management ,Economics, Econometrics and Finance (miscellaneous) ,Validity ,Feature selection ,Machine learning ,computer.software_genre ,lcsh:HG8011-9999 ,lcsh:Insurance ,Accounting ,random forest algorithms ,0502 economics and business ,ddc:330 ,Feature (machine learning) ,Bayesian regulatory neural network ,040101 forestry ,050208 finance ,Artificial neural network ,business.industry ,05 social sciences ,04 agricultural and veterinary sciences ,Random forest ,capital adequacy ,Capital adequacy ratio ,machine learning ,BIS capital adequacy ratio ,Ordinary least squares ,0401 agriculture, forestry, and fisheries ,Artificial intelligence ,business ,computer ,Algorithm - Abstract
The purpose of this study is to find the most important variables that represent the future projections of the Bank of International Settlements’ (BIS) capital adequacy ratio, which is the index of financial soundness in a bank as a comprehensive and important measure of capital adequacy. This study analyzed the past 12 years of data from all domestic banks in South Korea. The research data include all financial information, such as key operating indicators, major business activities, and general information of the financial supervisory service of South Korea from 2008 to 2019. In this study, machine learning techniques, Random Forest Boruta algorithms, Random Forest Recursive Feature Elimination, and Bayesian Regularization Neural Networks (BRNN) were utilized. Among 1929 variables, this study found 38 most important variables for representing the BIS capital adequacy ratio. An additional comparison was executed to confirm the statistical validity of future prediction performance between BRNN and ordinary least squares (OLS) models. BRNN predicted the BIS capital adequacy ratio more robustly and accurately than the OLS models. We believe our findings would appeal to the readership of your journal such as the policymakers, managers and practitioners in the bank-related fields because this study highlights the key findings from the data-driven approaches using machine learning techniques.
- Published
- 2021
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.