|
|
|
|
LEADER |
02956nmm a2200661 u 4500 |
001 |
EB002154601 |
003 |
EBX01000000000000001292727 |
005 |
00000000000000.0 |
007 |
cr||||||||||||||||||||| |
008 |
230404 ||| eng |
020 |
|
|
|a 9798400234828
|
100 |
1 |
|
|a Chan-Lau, Jorge
|
245 |
0 |
0 |
|a Surrogate Data Models: Interpreting Large-scale Machine Learning Crisis Prediction Models
|c Jorge Chan-Lau, Ruofei Hu, Maksym Ivanyna, Ritong Qu, Cheng Zhong
|
260 |
|
|
|a Washington, D.C.
|b International Monetary Fund
|c 2023
|
300 |
|
|
|a 31 pages
|
651 |
|
4 |
|a United States
|
653 |
|
|
|a Interest rates
|
653 |
|
|
|a Economics
|
653 |
|
|
|a Finance
|
653 |
|
|
|a Financial crises
|
653 |
|
|
|a Financial services
|
653 |
|
|
|a Yield curve
|
653 |
|
|
|a Prices, Business Fluctuations, and Cycles: Forecasting and Simulation
|
653 |
|
|
|a Deposit rates
|
653 |
|
|
|a Economics of specific sectors
|
653 |
|
|
|a Currency crises
|
653 |
|
|
|a Forecasting and Other Model Applications
|
653 |
|
|
|a Capital and Ownership Structure
|
653 |
|
|
|a Goodwill
|
653 |
|
|
|a Macroeconomics
|
653 |
|
|
|a Early warning systems
|
653 |
|
|
|a Financial Risk and Risk Management
|
653 |
|
|
|a Financing Policy
|
653 |
|
|
|a Financial Risk Management
|
653 |
|
|
|a Economic & financial crises & disasters
|
653 |
|
|
|a Machine learning
|
653 |
|
|
|a Technological Change: Choices and Consequences
|
653 |
|
|
|a Technology
|
653 |
|
|
|a Large Data Sets: Modeling and Analysis
|
653 |
|
|
|a Value of Firms
|
653 |
|
|
|a Economics: General
|
653 |
|
|
|a Informal sector
|
653 |
|
|
|a Crisis management
|
653 |
|
|
|a Diffusion Processes
|
653 |
|
|
|a Intelligence (AI) & Semantics
|
653 |
|
|
|a Banks and Banking
|
653 |
|
|
|a Interest Rates: Determination, Term Structure, and Effects
|
653 |
|
|
|a Financial Crises
|
700 |
1 |
|
|a Hu, Ruofei
|
700 |
1 |
|
|a Ivanyna, Maksym
|
700 |
1 |
|
|a Qu, Ritong
|
041 |
0 |
7 |
|a eng
|2 ISO 639-2
|
989 |
|
|
|b IMF
|a International Monetary Fund
|
490 |
0 |
|
|a IMF Working Papers
|
028 |
5 |
0 |
|a 10.5089/9798400234828.001
|
856 |
4 |
0 |
|u https://elibrary.imf.org/view/journals/001/2023/041/001.2023.issue-041-en.xml?cid=529723-com-dsp-marc
|x Verlag
|3 Volltext
|
082 |
0 |
|
|a 330
|
520 |
|
|
|a Machine learning models are becoming increasingly important in the prediction of economic crises. The models, however, use datasets comprising a large number of predictors (features) which impairs model interpretability and their ability to provide adequate guidance in the design of crisis prevention and mitigation policies. This paper introduces surrogate data models as dimensionality reduction tools in large-scale crisis prediction models. The appropriateness of this approach is assessed by their application to large-scale crisis prediction models developed at the IMF. The results are consistent with economic intuition and validate the use of surrogates as interpretability tools
|