Forecasting recovery rates on non-performing loans with machine learning
نویسندگان
چکیده
Abstract We compare the performance of a wide set regression techniques and machine-learning algorithms for predicting recovery rates on non-performing loans, using private database from European debt collection agency. find that rule-based such as Cubist, boosted trees, random forests perform significantly better than other approaches. In addition to loan contract specificities, predictors refer bank process — prior portfolio’s sale collector are also shown enhance forecasting performance. These variables, derived time series contacts defaulted clients client reimbursements bank, help all identify debtors with different repayment ability and/or commitment, in general those potential.
منابع مشابه
Dealing with non-performing loans of banks.PDF
A lot of confusion and misunderstanding has been created by several commentators on the issue of non-performing loans (NPLs) of banking system. They take the absolute amount of such loans at the current point of time and compare it with the quantum of such loans in October 1999 and make a hue and cry that the situation has deteriorated because the quantum of NPLs has gone up. Such a simplistic ...
متن کاملThe Impact of Provisioning Policies on Non-Performing Loans
The Central Bank has specific regulations, including asset classification guidelines and how to perform loan’s loss provisioning to oversee banks and credit institutions. The Central Bank seeks to improve the quality of the loan and reduce the amount of non-performing loans by reducing the revenues of the bank or credit institution through imposing fine on balance of each of the categories of n...
متن کاملFinancial Soundness Indicators with Emphasis on Non-performing Loans in Iran’s Banking System
This paper analyzes the financial soundness indicators in Iran`s banking system. In the first part it emphasis on asset quality measure by the non-performing loans ratio. The non-performing loans grew rapidly in last decade in Iran’s banking system and it reached higher than 25 percent of total loans in 2010. We conclude that NPLs increase have impact on real part of economy in the concept...
متن کاملAnalysis of non-performing loans of banks with the regional economic approach: unbalanced panel data method
The high volume of non-performing loans and its trend over time is one of the main problems of banks and central banking in Iran. This high volume of non-current receivables and the destructive effects that this phenomenon has, has made it very important to identify and examine the role of the factors affecting it. In the present study, determinant factors from the Borrower, lender and regional...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: International Journal of Forecasting
سال: 2021
ISSN: ['1872-8200', '0169-2070']
DOI: https://doi.org/10.1016/j.ijforecast.2020.06.009