Browse

You are looking at 1 - 2 of 2 items for :

  • Type: Journal Issue x
  • Intelligence (AI) & Semantics x
  • Industries; Land Use; Labor x
Clear All Modify Search
Mr. Jorge A Chan-Lau and Ran Wang
We introduce unFEAR, Unsupervised Feature Extraction Clustering, to identify economic crisis regimes. Given labeled crisis and non-crisis episodes and the corresponding features values, unFEAR uses unsupervised representation learning and a novel mode contrastive autoencoder to group episodes into time-invariant non-overlapping clusters, each of which could be identified with a different regime. The likelihood that a country may experience an econmic crisis could be set equal to its cluster crisis frequency. Moreover, unFEAR could serve as a first step towards developing cluster-specific crisis prediction models tailored to each crisis regime.
International Monetary Fund. Strategy, Policy, & and Review Department
"The first data and statistics strategy for the Fund comes at a critical time. A fast-changing data landscape, new data needs for evolving surveillance priorities, and persisting data weaknesses across the membership pose challenges and opportunities for the Fund and its members. The challenges emerging from the digital revolution include an unprecedented amount of new data and measurement questions on growth, productivity, inflation, and welfare. Newly available granular and high-frequency (big) data offer the potential for more timely detection of vulnerabilities. In the wake of the crisis, Fund surveillance requires greater cross-country data comparability; staff and authorities face the complexity of integrating new data sources and closing data gaps, while working to address the weaknesses noted by the IEO Report (Behind the Scenes with Data at the IMF) in 2016. The overarching strategy is to move toward an ecosystem of data and statistics that enables the Fund and its members to better meet the evolving data needs in a digital world. It integrates Fund-wide work streams on data provision to the Fund for surveillance purposes, international statistical standards, capacity development, and data management under a common institutional objective. It seeks seamless access and sharing of data within the Fund, enabling cloud-based data dissemination to support data provision by member countries (e.g., the “global data commons”), closing data gaps with new sources including Big Data, and improving assessments of data adequacy for surveillance to help better prioritize capacity development. The Fund also will work with policymakers to understand the implications of the digital economy and digital data for the macroeconomic statistics, including new measures of welfare beyond GDP."