scholarly journals Determining customer limits by data mining methods in credit allocation process

Author(s):  
Tuğçe Ayhan ◽  
Tamer Uçar

The demand for credit is increasing constantly. Banks are looking for various methods of credit evaluation that provide the most accurate results in a shorter period in order to minimize their rising risks. This study focuses on various methods that enable the banks to increase their asset quality without market loss regarding the credit allocation process. These methods enable the automatic evaluation of loan applications in line with the sector practices, and enable determination of credit policies/strategies based on actual needs. Within the scope of this study, the relationship between the predetermined attributes and the credit limit outputs are analyzed by using a sample data set of consumer loans. Random forest (RF), sequential minimal optimization (SMO), PART, decision table (DT), J48, multilayer perceptron(MP), JRip, naïve Bayes (NB), one rule (OneR) and zero rule (ZeroR) algorithms were used in this process. As a result of this analysis, SMO, PART and random forest algorithms are the top three approaches for determining customer credit limits.

Author(s):  
Tushar ◽  
Tushar ◽  
Shibendu Shekhar Roy ◽  
Dilip Kumar Pratihar

Clustering is a potential tool of data mining. A clustering method analyzes the pattern of a data set and groups the data into several clusters based on the similarity among themselves. Clusters may be either crisp or fuzzy in nature. The present chapter deals with clustering of some data sets using Fuzzy C-Means (FCM) algorithm and Entropy-based Fuzzy Clustering (EFC) algorithm. In FCM algorithm, the nature and quality of clusters depend on the pre-defined number of clusters, level of cluster fuzziness and a threshold value utilized for obtaining the number of outliers (if any). On the other hand, the quality of clusters obtained by the EFC algorithm is dependent on a constant used to establish the relationship between the distance and similarity of two data points, a threshold value of similarity and another threshold value used for determining the number of outliers. The clusters should ideally be distinct and at the same time compact in nature. Moreover, the number of outliers should be as minimum as possible. Thus, the above problem may be posed as an optimization problem, which will be solved using a Genetic Algorithm (GA). The best set of multi-dimensional clusters will be mapped into 2-D for visualization using a Self-Organizing Map (SOM).


Author(s):  
Knut Gerlach ◽  
Olaf Hübler

SummaryFirms are affected by the product demand. This leads to employment adjustments. In the literature we find only very few contributions investigating the issue whether internal adjustments are linked and which relationships exist with external adjustments. Are they of a complementary or substitutive nature? Furthermore, it is of interest to find out, whether we can observe an obvious trend and whether the adjustments are driven by cyclical movements.For this study we have an extensive data set of a large German manufacturing company, which supplies innovative products for the domestic and international market, provided on amonthly base from January 1999 to December 2005. The empirical analysis starts with descriptive statistics. We find that the employment adjustment cycle coincides only to a certain degree with the macroeconomic cycle. Internal and external adjustments are more characterized by complementarity than by substitution. Over the observed period we cannot detect analogous wage adjustments. It is noticeable that in 2003 compared with the years before the number of employees is substantially reduced. The econometric investigation is based on a two-stage approach. We start with a bivariate probit estimation in order to extract the relationship between the probability of overtime and of promotion. Unobserved variables have opposite effects on the former and the latter adjustment instrument. Furthermore, we detect a negative trend of internal employment adjustments. Cyclical effects are ambiguous. The next step, the determination of external adjustments with respect to overtime and promotion adjustments, is split into two estimates. On the one hand we do not distinguish between the type of external employment adjustment and on the other hand we use this information separating between quits, layoffs, workers with a cancellation agreement and with a transition into a transfer organisation. The first approach demonstrates that a promotion reduces the probability to leave the firm while overtime is positively associated with an external job change. This pattern holds generally speaking in the second, more detailed estimates. Quits are the exception. In this case we observe opposite effects. Finally, we cannot detect any influences of promotions on cancellation agreements.


2010 ◽  
Vol 10 (1) ◽  
pp. 1-25 ◽  
Author(s):  
H. Guan ◽  
R. Esswein ◽  
J. Lopez ◽  
R. Bergstrom ◽  
A. Warnock ◽  
...  

Abstract. We have quantified the relationship between Aerosol Index (AI) measurements and plume height for young biomass burning plumes using coincident OMI and CALIPSO measurements. This linear relationship allows the determination of high-altitude plumes wherever AI data are available, and it provides a data set for validating global fire plume injection heights in chemistry transport models. We find that all plumes detected from June 2006 to February 2009 with an AI value ≥9 are located at altitudes higher than 5 km. Older high-altitude plumes have lower AI values than young plumes at similar altitudes. We have examined available AI data from the OMI and TOMS instruments (1978–2009) and find that large AI plumes occur more frequently over North America than over Australia or Russia/Northeast Asia. According to the derived relationship, during this time interval, 181 plumes reached altitudes above 8 km. One hundred and thirty-two had injection heights ≥8 km but below 12 km, and 49 were lofted to 12 km or higher, including 14 plumes injected above 16 km.


2022 ◽  
Vol 14 (2) ◽  
pp. 798
Author(s):  
Snezhana Gocheva-Ilieva ◽  
Atanas Ivanov ◽  
Maya Stoimenova-Minova

A novel framework for stacked regression based on machine learning was developed to predict the daily average concentrations of particulate matter (PM10), one of Bulgaria’s primary health concerns. The measurements of nine meteorological parameters were introduced as independent variables. The goal was to carefully study a limited number of initial predictors and extract stochastic information from them to build an extended set of data that allowed the creation of highly efficient predictive models. Four base models using random forest, CART ensemble and bagging, and their rotation variants, were built and evaluated. The heterogeneity of these base models was achieved by introducing five types of diversities, including a new simplified selective ensemble algorithm. The predictions from the four base models were then used as predictors in multivariate adaptive regression splines (MARS) models. All models were statistically tested using out-of-bag or with 5-fold and 10-fold cross-validation. In addition, a variable importance analysis was conducted. The proposed framework was used for short-term forecasting of out-of-sample data for seven days. It was shown that the stacked models outperformed all single base models. An index of agreement IA = 0.986 and a coefficient of determination of about 95% were achieved.


Author(s):  
Beriliana Hapsari ◽  
R. Nelly Nur Apandi

The use of fair value method has positive and negative impact. The positive impact of applying the fair value method is more relevant for the decision maker and shows the economic value according to the circumstances at that time. While its negative impact, can lead to manipulation and uncertainty. It takes greater effort for the auditor to assess fairness of fair non-current asset quality, resulting in increased audit fees. This study aims to determine the effect of fair value of non-current asset on the determination of audit fees and to know the multiple large shareholder moderation in the relationship between fair value of non-current asset and audit fee. Using the OLS regression, this study uses companies listed on the IDX from 2013 to 2015 with the exception of the financial sector. The result of this research indicates that fair value of non-current asset influences audit fee and this research shows that the second largest ownership can weaken the effect of fair value of non-current asset toward  audit fees.


2016 ◽  
Vol 17 (01) ◽  
pp. 1-14
Author(s):  
Mpa Saputra

The Objective of the research is to determine the relationship between the level of education the community Bojong Rawa village and environmental responsibility with the participation of the health maintenance environment. The method which is used in this research is the correlation associative quantitative approach of 60 people a total sample. Data were collected through questionnaires. Analysis of data to test the hypothesis of research is used the regression analysis. The research concluded that; First, there is a positive relationship between the level of education with the participation of the health maintenance environment with the correllation coefficient of  0, 86 at  = 0, 05 on  = 39, 71 + 3, 191 X1 the regression equation. Second, have a positive relationship between environmental responsibility and participation of the healthmaintenance environment with the correlation  regression equation. Third, there is a positive relationship between the level of education and environmental responsibility together with the participation of the health maintenance environment with multiple  correlation coefficient () of () with coefficients of determination of 96 %. The double regression equation   = - 0, 7 + - 0, 64 + 1, 13.


2019 ◽  
Vol 16 (3) ◽  
pp. 292-306
Author(s):  
Jana Kotěšovcová ◽  
Jiří Mihola ◽  
Petr Budinský

The sovereign credit rating provides information about the creditworthiness of a country and thereby serves as a tool for investors in order to make right decisions concerning financial assets worth investments. Thus, determination of a sovereign credit rating is a highly complex and challenging activity. Specialized agencies are involved in rating assessment. So, it’s essential to analyze the efficiency of their work and seek out easily accessible tools for generating assessments of such ratings. The objective of this article is to find out whether sovereign credit rating can be reliably estimated using trends of selected macroeconomic indicators, despite the fact that sovereign credit rating is most likely influenced by non-economic factors. This can be used for strategic considerations at national and multinational levels. The relationships between sovereign credit rating and the trends of macroeconomic indicators were examined using statistical methods, linear multiple regression analysis, cumulative correlation coefficient, and multicollinearity test. The data source used is comprised of selected World Bank indicators meeting the conditions of completeness and representativeness. The data set has shown a cumulative correlation coefficient value greater than 95%, however at 100% multicollinearity. This is followed by the gradual elimination of indicators, but even this did not allow achieving acceptable values. So, the conclusion is that rating levels are not explainable solely by the trends of economic indicators, but other influences, e.g. political. However, the fact that the statistical model yielded acceptable results for five and fewer indicators allowed a regression equation to be found that gives good estimates of a country’s rating. This allows, for example, predicting of ratings relatively easy by forecasting the development of selected macroeconomic indicators.


2001 ◽  
Vol 34 (1) ◽  
pp. 42-46 ◽  
Author(s):  
Paul F. Henry ◽  
Mark T. Weller ◽  
Chick C. Wilson

The use of isotopes to extract precise and very accurate structural information in complex Rietveld analysis has been demonstrated by comparison of results obtained for single-data-set analysis with those from multiple-data-set single-crystallographic-model analysis of NixMg1−xO doped with various nickel isotopes. Fractional occupancies of dopants can be accurately determined down to at least the 0.5% doping level, which cannot be obtained through single-sample-data-set refinements because of correlation effects, even when a large contrast exists between the nickel isotope used and magnesium.


1994 ◽  
Vol 72 (03) ◽  
pp. 426-429 ◽  
Author(s):  
S Kitchen ◽  
I D Walker ◽  
T A L Woods ◽  
F E Preston

SummaryWhen the International Normalised Ratio (INR) is used for control of oral anticoagulant therapy the same result should be obtained irrespective of the laboratory reagent used. However, in the UK National External Quality Assessment Scheme (NEQAS) for Blood Coagulation INRs determined using different reagents have been significantly different.For 18 NEQAS samples Manchester Reagent (MR) was associated with significantly lower INRs than those obtained using Diagen Activated (DA, p = 0.0004) or Instrumentation Laboratory PT-Fib HS (IL, p = 0.0001). Mean INRs for this group were 3.15, 3.61, and 3.65 for MR, DA, and IL respectively. For 61 fresh samples from warfarin-ised patients with INRs of greater than 3.0 the relationship between thromboplastins in respect of INR was similar to that observed for NEQAS data. Thus INRs obtained with MR were significantly lower than with DA or IL (p <0.0001). Mean INRs for this group were 4.01, 4.40, and 4.59 for MR, DA, and IL respectively.We conclude that the differences between INRs measured with the thromboplastins studied here are sufficiently great to influence patient management through warfarin dosage schedules, particularly in the upper therapeutic range of INR. There is clearly a need to address the issues responsible for the observed discrepancies.


2016 ◽  
pp. 137-142
Author(s):  
V.O. Benyuk ◽  
◽  
V.M. Goncharenko ◽  
T.R. Nykoniuk ◽  
◽  
...  

The objective: to еxplore the relationship between the activity of endometrial proliferation and the state of the local immune response in the uterus in the conditions berprestasi process. Patients and methods. Examined 228 women of reproductive and perimenopausal age with endometrial pathology using ultrasound and then performing hysteroresectoscopy. Determination of the concentrations of the cytokines IL-1, IL-2, IL-6 and TNF was performed by solid phase ELISA. Results. Found a trend that confirms the loss of sensitivity to hormones at the stage of malignancy of the endometrium and can be used as diagnostic determinants in determining the nature of intrauterine pathology and criterion of the effectiveness of conservative therapy. Conclusion. Improving etiopatogenetice approach to the therapy of hyperplastic proce.sses of endometrium with determination of receptor phenotype of the endometrium is a research direction in modern gynecology, which will help to improve the results of treatment and prevention of intrauterine pathology. Key words: endometrial hyperplasia,the receptors for progesterone and estrogen, immunohistochemical method.


Sign in / Sign up

Export Citation Format

Share Document