|
on Forecasting |
| By: | Tae-Hwy Lee (Department of Economics, University of California Riverside); Ekaterina Seregina (Colby College) |
| Abstract: | In this paper we develop a novel method of combining many forecasts based on a machine learning algorithm called Graphical LASSO. We visualize forecast errors from different forecasters as a network of interacting entities and generalize network inference in the presence of common factor structure and structural breaks. First, we note that forecasters often use common information and hence make common mistakes, which makes the forecast errors exhibit common factor structures. We propose the Factor Graphical LASSO (Factor GLASSO), which separates common forecast errors from the idiosyncratic errors and exploits sparsity of the precision matrix of the latter. Second, since the network of experts changes over time as a response to unstable environments such as recessions, it is unreasonable to assume constant forecast combination weights. Hence, we propose Regime-Dependent Factor Graphical LASSO (RD-Factor GLASSO) and develop its scalable implementation using the Alternating Direction Method of Multipliers (ADMM) to estimate regime-dependent forecast combination weights. The empirical application to forecasting macroeconomic series using the data of the European Central Bank’s Survey of Professional Forecasters (ECB SPF) demonstrates superior performance of a combined forecast using Factor GLASSO and RD-Factor GLASSO. |
| Keywords: | Common Forecast Errors, Regime Dependent Forecast Combination, Sparse Precision Matrix of Idiosyncratic Errors, Structural Breaks. |
| JEL: | C13 C38 C55 |
| Date: | 2022–09 |
| URL: | http://d.repec.org/n?u=RePEc:ucr:wpaper:202213&r= |
| By: | Arai, Natsuki; Iizuka, Nobuo; Yamamoto, Yohei |
| Abstract: | This paper evaluates the efficiency of the Japanese fiscal authority’s revenue projections from 1960 to 2020 using real-time data. Revenue projections are not efficient, primarily due to the conditioning projections of output growth. By adjusting the forecasts based on the results of real-time forecast evaluations, this paper finds that the out-of-sample accuracy of the one-year-ahead projections could be significantly improved by a magnitude of up to 10 percent in root mean squared errors. The analysis of the disaggregated series suggests that corporate tax projections are the least efficient. The fiscal authority’s loss function is estimated to be asymmetric, making the underprediction of revenues more common. |
| Keywords: | Revenue Projections, Japan, Forecast Evaluation, Out-of-Sample Forecast Accuracy |
| JEL: | C53 E62 H68 |
| Date: | 2022–09 |
| URL: | http://d.repec.org/n?u=RePEc:hit:hiasdp:hias-e-122&r= |
| By: | Martin Casta |
| Abstract: | This paper presents a simple reduced-form error correction model for forecasting nominal exchange rates. The model is inspired by the classical monetary model of exchange rates. However, the commonly used monetary aggregates were replaced by loans to corporations. The reason for this change is that our goal is to focus on corporate deposits, for which corporate loans act as a proxy. For presentational purposes, we focus on eight major trading currency pairs: AUD/USD, CAD/USD, CHF/USD, EUR/USD, GBP/USD, NZD/USD, SEK/USD and JPY/USD, for which we use data from approximately the last two decades. We empirically show statistically and economically significant exchange rates forecastability in the medium and long run, and we also present some findings on predictability even in the short run. In short, our results suggest that corporate loans are a significant driver behind exchange rate movements. |
| Keywords: | Exchange rates, forecasting, forecast evaluation |
| JEL: | C5 F31 F32 F37 |
| Date: | 2022–08 |
| URL: | http://d.repec.org/n?u=RePEc:cnb:wpaper:2022/7&r= |