|
on Forecasting |
| By: | Klaus-Peter Hellwig |
| Abstract: | I regress real GDP growth rates on the IMF’s growth forecasts and find that IMF forecasts behave similarly to those generated by overfitted models, placing too much weight on observable predictors and underestimating the forces of mean reversion. I identify several such variables that explain forecasts well but are not predictors of actual growth. I show that, at long horizons, IMF forecasts are little better than a forecasting rule that uses no information other than the historical global sample average growth rate (i.e., a constant). Given the large noise component in forecasts, particularly at longer horizons, the paper calls into question the usefulness of judgment-based medium and long-run forecasts for policy analysis, including for debt sustainability assessments, and points to statistical methods to improve forecast accuracy by taking into account the risk of overfitting. |
| Date: | 2018–12–07 |
| URL: | http://d.repec.org/n?u=RePEc:imf:imfwpa:18/260&r=all |
| By: | Pönkä, Harri; Stenborg, Markku |
| Abstract: | We employ probit models to study the predictability of recession periods in Finland using a set of commonly used variables based on previous literature. The findings point out that individual predictors, including the term spread and the real housing prices from the capital area, are useful predictors of recession periods. However, the best in-sample fit is found using combinations of variables. The pseudo out-of-sample forecasting results are generally in line with the in-sample results, and suggest that in the one-quarter ahead forecasts a model combining the term spread, the unemployment expectation component of the consumer confidence index, and the consumer confidence index performs the best based on the area under the receiver operating characteristic curve. An autoregressive specification improves the in-sample fit of the models compared to the static probit model, but findings from pseudo out-of-sample forecasts vary between forecasting horizons. |
| Keywords: | Business cycle, Recession period, Probit model |
| JEL: | C22 E32 E37 |
| Date: | 2018–10–12 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:91226&r=all |
| By: | Luca Di Bonaventura; Mario Forni; Francesco Pattarin |
| Abstract: | We present a comparative analysis of the forecasting performance of two dynamic factor models, the Stock and Watson (2002a, b) model and the Forni, Hallin, Lippi and Reichlin (2005) model, based on vintage data. Our dataset that contains 107 monthly US “first release” macroeconomic and financial vintage time series, spanning the 1996:12 to 2017:6 period with monthly periodicity, extracted from the Bloomberg database†. We compute real-time one-month-ahead forecasts with both models for four key macroeconomic variables: the month-on-month change in industrial production, the unemployment rate, the core consumer price index and the ISM Purchasing Managers’ Index. First, we find that both the Stock and Watson and the Forni, Hallin, Lippi and Reichlin models outperform simple autoregressions for industrial production, unemployment rate and consumer prices, but that only the first model does so for the PMI. Second, we find that neither models always outperform the other. While Forni, Hallin, Lippi and Reichlin’s beats Stock and Watson’s in forecasting industrial production and consumer prices, the opposite happens for the unemployment rate and the PMI. |
| Keywords: | Dynamic factor models, Forecasting, Forecasting Performance, Vintage data,First release data |
| JEL: | C01 C32 C52 C53 E27 E37 |
| Date: | 2018–07 |
| URL: | http://d.repec.org/n?u=RePEc:mod:wcefin:0070&r=all |
| By: | Konstantin Styrin (Bank of Russia, Russian Federation) |
| Abstract: | In this study, I forecast CPI inflation in Russia by the method of Dynamic Model Averaging (Raftery et al., 2010; Koop and Korobilis, 2012) pseudo out-of-sample on historical data. This method can be viewed as an extension of the Bayesian Model Averaging where the identity of a model that generates data and model parameters are allowed to change over time. The DMA is shown not to produce forecasts superior to simpler benchmarks even if a subset of individual predictors is pre-selected “with the benefit of hindsight” on the full sample. The two groups of predictors that feature the highest average values of the posterior inclusion probability are loans to non-financial firms and individuals along with actual and anticipated wages. |
| Keywords: | Bayesian model averaging, model uncertainty, econometric modeling, high-dimension model, inflation forecast. |
| JEL: | C5 C53 E37 |
| Date: | 2018–12 |
| URL: | http://d.repec.org/n?u=RePEc:bkr:wpaper:wps39&r=all |
| By: | Franses, Ph.H.B.F.; Welz, M. |
| Abstract: | There are various reasons why professional forecasters may disagree in their quotes for macroeconomic variables. One reason is that they target at different vintages of the data. We propose a novel method to test forecast bias in case of such heterogeneity. The method is based on Symbolic Regression, where the variables of interest become interval variables. We associate the interval containing the vintages of data with the intervals of the forecasts. An illustration to 18 years of forecasts for annual USA real GDP growth, given by the Consensus Economics forecasters, shows the relevance of the method. |
| Keywords: | Forecast bias, Data revisions, Interval data, Symbolic regression |
| JEL: | C53 |
| Date: | 2019–09–01 |
| URL: | http://d.repec.org/n?u=RePEc:ems:eureir:114113&r=all |