Search Results

You are looking at 1 - 5 of 5 items for :

  • "backtesting performance" x
Clear All
International Monetary Fund. European Dept.

1. A Short Introduction on ML Algorithms 2. Backtesting Performance of Selected Methods APPENDIX I. Data Description CORPORATE LIQUIDITY AND SOLVENCY DURING THE PANDEMIC AND POLICY RESPONSE A. Introduction B. Data and Methodology C. Liquidity and Solvency Gaps Estimates D. Conclusions and Policy Response Going Forward References FIGURES 1. Corporate Sector: Pre-COVID Selected Indicators 2. Firms Distribution by Sector 3. Corporate Sector: Distribution of Firms by Liquidity and Solvency Stance 4. Corporate Sector: Liquidity and Solvency

International Monetary Fund. European Dept.

accurate during the first subperiod (2012:Q4–2015:Q4), likely due to the factor that this subperiod shows a clear upward trend in GDP growth, therefore an AR(1) with more weight (large coefficient) on the lagged variable is not an unreasonable guess. Table 2. Malta: Backtesting Performance of Selected Methods (RMSE, percentage points) Full sample Sub sample Methods 2012:Q4–2021:Q1 2012:Q4–2015:Q4 2016:Q1–2019:Q4 2020:Q1–2021:Q1 DFM 2.8 2.6 2.1 4.9 ML Algorithms Lasso 3.9 2.7 3.3 7

Mr. Jean-Francois Dauphin, Mr. Kamil Dybczak, Morgan Maneely, Marzie Taheri Sanjani, Mrs. Nujin Suphaphiphat, Yifei Wang, and Hanqi Zhang

given nowcasting method combines n such out-of-sample nowcasts for t =t’, t’-1, ..., t’-n+1, where n depends on historical data availability and the user’s specification. Finally, the Nowcasting Tracker produces actual nowcasts for the current quarter using all or some of the methods tested in the backtest. At this point the toolkit has completed a successful execution. A table and a plot of GDP nowcasts are generated and saved for the user’s information. The user can base their comprehensive judgement upon each method’s backtest performance as well as their

Mr. Jean-Francois Dauphin, Mr. Kamil Dybczak, Morgan Maneely, Marzie Taheri Sanjani, Mrs. Nujin Suphaphiphat, Yifei Wang, and Hanqi Zhang
This paper describes recent work to strengthen nowcasting capacity at the IMF’s European department. It motivates and compiles datasets of standard and nontraditional variables, such as Google search and air quality. It applies standard dynamic factor models (DFMs) and several machine learning (ML) algorithms to nowcast GDP growth across a heterogenous group of European economies during normal and crisis times. Most of our methods significantly outperform the AR(1) benchmark model. Our DFMs tend to perform better during normal times while many of the ML methods we used performed strongly at identifying turning points. Our approach is easily applicable to other countries, subject to data availability.