oalib
Search Results: 1 - 10 of 100 matches for " "
All listed articles are free for downloading (OA Articles)
Page 1 /100
Display every page Item
Logistic and SVM Credit Score Models Based on Lasso Variable Selection  [PDF]
Qingqing Li
Journal of Applied Mathematics and Physics (JAMP) , 2019, DOI: 10.4236/jamp.2019.75076
Abstract:
There are many factors influencing personal credit. We introduce Lasso technique to personal credit evaluation, and establish Lasso-logistic, Lasso-SVM and Group lasso-logistic models respectively. Variable selection and parameter estimation are also conducted simultaneously. Based on the personal credit data set from a certain lending platform, it can be concluded through experiments that compared with the full-variable Logistic model and the stepwise Logistic model, the variable selection ability of Group lasso-logistic model was the strongest, followed by Lasso-logistic and Lasso-SVM respectively. All three models based on Lasso variable selection have better filtering capability than stepwise selection. In the meantime, the Group lasso-logistic model can eliminate or retain relevant virtual variables as a group to facilitate model interpretation. In terms of prediction accuracy, Lasso-SVM had the highest prediction accuracy for default users in the training set, while in the test set, Group lasso-logistic had the best classification accuracy for default users. Whether in the training set or in the test set, the Lasso-logistic model has the best classification accuracy for non-default users. The model based on Lasso variable selection can also better screen out the key factors influencing personal credit risk.
Intelligence system of supply chain management of logistic company based on the discrete event, agent and system dynamic simulation models  [PDF]
Ganyukov Vladimir Yurievich,Khanova Anna Alekseevna,Suldina Nataliya Viсtorovna
Vestnik Astrahanskogo Gosudarstvennogo Tehni?eskogo Universiteta. Seria: Upravlenie, Vy?islitel?naa Tehnika i Informatika , 2012,
Abstract: Simulation modeling and intelligence analysis of the sales are effective tools for the analysis of logistic network management of the building company. The technology of management of logistic chains based on the multi-variant simulation models will automate the management process by designing the structure of the logistics network. Exact demand forecasting, plan collaboration and control of the processes in the logistic network, the operational decision-making will be provided by the system of the intelligence data analysis. The developed technology of the analysis of supply chain management system will help to realize the tasks of improving the efficiency of the operation by providing tools of the analysis of changes within the logistic network and in market environment, as well as of adaptive planning and coordination of the processes for all participants of the logistic network.
Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data
Behzad Eftekhar, Kazem Mohammad, Hassan Ardebili, Mohammad Ghodsi, Ebrahim Ketabchi
BMC Medical Informatics and Decision Making , 2005, DOI: 10.1186/1472-6947-5-3
Abstract: 1000 Logistic regression and ANN models based on initial clinical data related to the GCS, tracheal intubation status, age, systolic blood pressure, respiratory rate, pulse rate, injury severity score and the outcome of 1271 mainly head injured patients were compared in this study. For each of one thousand pairs of ANN and logistic models, the area under the receiver operating characteristic (ROC) curves, Hosmer-Lemeshow (HL) statistics and accuracy rate were calculated and compared using paired T-tests.ANN significantly outperformed logistic models in both fields of discrimination and calibration but under performed in accuracy. In 77.8% of cases the area under the ROC curves and in 56.4% of cases the HL statistics for the neural network model were superior to that for the logistic model. In 68% of cases the accuracy of the logistic model was superior to the neural network model.ANN significantly outperformed the logistic models in both fields of discrimination and calibration but lagged behind in accuracy. This study clearly showed that any single comparison between these two models might not reliably represent the true end results. External validation of the designed models, using larger databases with different rates of outcomes is necessary to get an accurate measure of performance outside the development population.In recent years, outcome prediction studies have become the avante garde in many areas of health care research, especially in critical care and trauma. However acceptable models for outcome prediction have been difficult to develop [1]. According to Wyatt and Altman, to be useful, a predictive model must be simple to calculate, have an apparent structure and be tested in independent data sets with evidence of generality [2]. While this is a high standard, availability and popularity of portable computers, deprioritize the need for simplicity of the model and having an apparent structure.Artificial neural networks (ANNs) are mathematical constructs m
The Quasi cellular nets-based models of transport and logistic systems  [PDF]
Anton Aristov
Computer Science , 2015,
Abstract: There are many systems in different subjects such as industry, medicine, transport, social and others, can be discribed on their dynamic of flows. Nowadays models of flows consist of micro- and macro-models. In practice there is a problem of convertation from different levels of simulation. In the different articles author descriptes quasi cellular nets. Quasi cellular nets are new type of discrete structures without signature. It may be used for simulation instruments. This structures can simulate flows on micro- and macro levels on the single model structure. In this article described using quasi cellular nets in transport and logistics of open-cast mining.
Goodness of fit of logistic models for random graphs  [PDF]
Pierre Latouche,Stéphane Robin,Sarah Ouadah
Statistics , 2015,
Abstract: Logistic models for random graphs are commonly used to study binary networks when covariate information is available. After estimating the logistic parameters, one of the main questions which arises in practice is to assess the goodness of fit of the corresponding model. To address this problem, we add a general term, related to the graphon function of W-graph models, to the logistic function. Such an extra term aims at characterizing the residual structure of the network, that is not explained by the covariates. We approximate this new generic logistic model using a class of models with blockwise constant residual structure. This framework allows to derive a Bayesian procedure from a model based selection context using goodness-of-fit criteria. All these criteria depend on marginal likelihood terms for which we do provide estimates relying on two series of variational approximations. Experiments on toy data are carried out to assess the inference procedure. Finally, two real networks from social sciences and ecology are studied to illustrate the proposed methodology.
Identifiability of multivariate logistic mixture models  [PDF]
ZiQiang Shi,TieRan Zheng,JiQing Han
Mathematics , 2012,
Abstract: Mixture models have been widely used in modeling of continuous observations. For the possibility to estimate the parameters of a mixture model consistently on the basis of observations from the mixture, identifiability is a necessary condition. In this study, we give some results on the identifiability of multivariate logistic mixture models.
A note on logistic regression and logistic kernel machine models  [PDF]
Ru Wang,Jie Peng,Pei Wang
Statistics , 2011,
Abstract: This is a note on logistic regression models and logistic kernel machine models. It contains derivations to some of the expressions in a paper -- SNP Set Analysis for Detecting Disease Association Using Exon Sequence Data -- submitted to BMC proceedings by these authors.
Evaluating software architecture using fuzzy formal models  [PDF]
Payman Behbahaninejad,Ali Harounabadi,Sayed Javad Mirabedini
Management Science Letters , 2012,
Abstract: Unified Modeling Language (UML) has been recognized as one of the most popular techniques to describe static and dynamic aspects of software systems. One of the primary issues in designing software packages is the existence of uncertainty associated with such models. Fuzzy-UML to describe software architecture has both static and dynamic perspective, simultaneously. The evaluation of software architecture design phase initiates always help us find some additional requirements, which helps reduce cost of design. In this paper, we use a fuzzy data model to describe the static aspects of software architecture and the fuzzy sequence diagram to illustrate the dynamic aspects of software architecture. We also transform these diagrams into Petri Nets and evaluate reliability of the architecture. The web-based hotel reservation system for further explanation has been studied.
Alternative Goodness-of-Fit Test in Logistic Regression Models
M.E. Nja,E.I. Enang,A.U. Chukwu,C.A. Udomboso
Journal of Modern Mathematics and Statistics , 2012, DOI: 10.3923/jmmstat.2011.43.46
Abstract: The Deviance and the Pearson chi-square are two traditional goodness-of-fit tests in generalized linear models for which the logistic model is a special case. The effort involved in the computation of either the Deviance or Pearson chi-square statistic is enormous and this provides a reason for prospecting an alternative goodness-of-fit test in logistic regression models with discrete predictor variables. The Deviance is based on the log likelihood function while the Pearson chi-square derives from the discrepancies between observed and predicted counts. Replacing observed and predicted counts with observed proportions and predicted probabilities, respectively in a cross-classification data arrangement, the standard error of estimate is proposed as an alternative goodness-of-fit test in logistic regression models. The illustrative example returns favourable comparisons with Deviance and the Pearson chi-square statistics.
Two logistic models for the prediction of hypothyroidism in pregnancy
Anthony U Mbah, Emmanuel C Ejim, Obinna D Onodugo, Francis O Ezugwu, Matthew I Eze, Peter O Nkwo, Winston C Ugbajah
BMC Research Notes , 2011, DOI: 10.1186/1756-0500-4-205
Abstract: The study was conducted in two phases. The phase one study comprised of healthy women in different stages of pregnancy who attended routine antenatal clinic at St Theresa's Maternity Hospital, Enugu, Nigeria from September 6 to October 18 1994. In this study the variables compared between the hypothyroid and non-hypothyroid pregnant women were maternal age, the number of the pregnancy or gravidity, gestational age, social class, body weight, height, the clinically assessed size of the thyroid gland, serum free thyroxin (FT4) and serum thyrotrophin (TSH). Based on the parameter differences between the two comparison groups of pregnant women two Logistic models, Model I and Model 11, were derived to differentiate the hypothyroid group from their non-hypothyroid counterparts. The two logistic models were then applied in a prospective validation study involving 197 pregnant women seen at presentation in Mother of Christ Specialist Hospital and Maternity, Ogui Road, Enugu from March 2002 to November 2007The findings were that 82 (50.3%) of the 163 pregnant women had thyroid gland enlargement while 60 (36.8%) had hypothyroidism as defined by FT4 values below and/or TSH above their laboratory reference ranges. The pregnant subjects with hypothyroidism, compared with their non-hypothyroid counterparts, were characterized by a higher gravidity (p < 0.01), a higher body weight (p < 0.01), a higher goiter prevalence rate (p < 0.01) and a more advanced gestational age (p < 0.0001). A significant, positive correlation was also found between body weight and gestational age (r = 0.5; p < 0.01) At the cut-off point for Model l (fitted with gravidity, thyroid size and gestational age) it had a sensitivity of 100%, a specificity of 72.8% and an overall predictive accuracy of 82.9%; whereas for Model II (fitted with gravidity, thyroid size and body weight) the sensitivity was 100%, the specificity was 59.2% and the overall accuracy of discrimination was 74.8%. In the prospective valid
Page 1 /100
Display every page Item


Home
Copyright © 2008-2017 Open Access Library. All rights reserved.