Discriminant Analysis of Zero Recovery for China's NPL

Classification of whether recovery of non-performing loans (NPL) is zero or positive is not only important in management of non-performing loans, but also is essential for estimating recovery rate and implementing the new Basel Capital Accord. Based on the largest database of NPL's recovering information in China, this paper tries to establish discriminant models to predict the loan with zero recovery. We first use Step-wise discrimination method to select variables; then give an in-depth analysis on why the selected variables are important factors influencing whether a loan is zero or positive recovery rate. Using the selected variables, we establish two-type discriminant models to classify the NPLs. Empirical results show that both models achieve high prediction accuracy, and the characteristics of obligors are the most important factors in determining whether a NPL is positively recovered or zero recovered.

Download Full-text

Perbandingan Analisis Diskriminan dan Analisis Regresi Logistik Ordinal dalam Prediksi Klasifikasi Kondisi Kesehatan Bank

Jurnal Matematika ◽

10.24843/jmat.2017.v07.i02.p86 ◽

2017 ◽

Vol 7 (2) ◽

pp. 92

Author(s):

Fajri Zufa ◽

Sigit Nugroho ◽

Mudin Simanihuruk

Keyword(s):

Logistic Regression ◽

Regression Analysis ◽

Discriminant Analysis ◽

Logistic Regression Analysis ◽

Prediction Accuracy ◽

Secondary Data ◽

Ordinal Logistic Regression ◽

Capital Adequacy ◽

Ordinal Logistic Regression Analysis

The purpose of this research is to compare the accuracy of bank classification prediction based on Capital Adequacy Ratio (CAR), Earning Asset Quality (EAQ), Non Performing Loan (NPL), Return on Assets (ROA), Net Interest Margin (NIM), Short Term Mismatch (STM) and Loan to Deposit Ratio (LDR). Discriminant analysis and ordinal logistic regression analysis are compared in classifying the prediction. The data used are secondary data, namely data classification of bank conditions in Indonesia in 2014 obtained from research institute PT Infovesta Utama. Based on Apparent Error Rate (APER) score obtained, it can be said that discriminant analysis is better in predicting the classification of bank conditions in Indonesia than that of ordinal logistic regression analysis. Discriminant analysis has the average prediction accuracy of 80%, while ordinal logistic regression analysis has the average prediction accuracy of 74,38%.

Download Full-text

BDA Models and Application for Recognizing the Headstreams of Mine Waterburst

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.34-35.1788 ◽

2010 ◽

Vol 34-35 ◽

pp. 1788-1793

Author(s):

W. Wan ◽

L. Jie

Keyword(s):

Discriminant Analysis ◽

Prediction Accuracy ◽

Cross Validation ◽

Data Sources ◽

Validation Method ◽

High Prediction Accuracy ◽

High Prediction ◽

Bayes Discriminant Analysis

BAYES discriminant analysis (BDA) method was used in the study of headstreams of prediction of minewater inrush and two BDA models for recognizing two-headstreams and multi-headstreams were constructed. Baesd on the principle of BDA theory and the classical headstream samples of different mines, the discriminant process and cross-validation method were introduced. 10 samples from a mine of HUA BEI Mine and 39 samples of JIAO ZUO Mine were used as data sources. Ca2+, Mg2+, Na+, K+, Cl-, HCO3-, SO42-, NO32-, F- and pH were selected as discriminant genes for two-headstreams BDA model and Na++K+, Ca2+, Mg2+, Cl-, SO42-, HCO3- were regarded as discriminant genes for multi-headstreams BDA model. Compared with the results of SQT method, ANN method and SVM method, the results show that the frame of BDA model was steady and high prediction accuracy can be obtained. BDA method and can be used in practical mine engineering.

Download Full-text

Discriminant analysis and its role in the classification of airmen.

PsycEXTRA Dataset ◽

10.1037/e402932004-001 ◽

1957 ◽

Cited By ~ 1

Author(s):

John A. Creager

Keyword(s):

Discriminant Analysis

Download Full-text

Classification of electricity load forecasting based on the factors influencing the load consumption and methods used: An-overview

2015 IEEE Conference on Energy Conversion (CENCON) ◽

10.1109/cencon.2015.7409585 ◽

2015 ◽

Cited By ~ 10

Author(s):

M. Mustapha ◽

M. W. Mustafa ◽

S. N Khalid ◽

I. Abubakar ◽

H. Shareef

Keyword(s):

Load Forecasting ◽

Factors Influencing ◽

Electricity Load ◽

Electricity Load Forecasting

Download Full-text

Small Bowel Obstruction

AACN Advanced Critical Care ◽

10.4037/15597768-2010-2009 ◽

2010 ◽

Vol 21 (2) ◽

pp. 187-194

Author(s):

Colleen Trevino

Keyword(s):

Small Bowel ◽

Physical Examination ◽

Surgical Intervention ◽

Signs And Symptoms ◽

Bowel Ischemia ◽

Key Factors ◽

The Past ◽

Factors Influencing ◽

Imaging Response

Strategies for the management of small bowel obstructions have changed significantly over the years. Nonoperative medical management has become the mainstay of treatment of many small bowel obstructions. However, the key to the management of small bowel obstructions is identifying those patients who need surgical intervention. Identification of those at risk for bowel ischemia and bowel death is an art as much as it is a science. Using the current literature and the past knowledge regarding small bowel obstructions, the clinician must carefully identify the signs and symptoms that suggest the need for operative intervention. Classification of the obstruction, history and physical examination, imaging, response to decompression and resuscitation, and resolution or progression of symptoms are the key factors influencing the management of small bowel obstructions.

Download Full-text