|
on Forecasting |
| By: | Faria, Gonçalo; Verona, Fabio |
| Abstract: | Predictability is time and frequency dependent. We propose a new forecasting method - forecast combination in the frequency domain - that takes this fact into account. With this method we forecast the equity premium and real GDP growth rate. Combining forecasts in the frequency domain produces markedly more accurate predictions relative to the standard forecast combination in the time domain, both in terms of statistical and economic measures of out-of-sample predictability. In a real-time forecasting exercise, the flexibility of this method allows to capture remarkably well the sudden and abrupt drops associated with recessions and further improve predictability. |
| Keywords: | forecast combination, frequency domain, equity premium, GDP growth, Haar filter, wavelets |
| JEL: | C58 G11 G17 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:bofrdp:12023&r=for |
| By: | Santiago Etchegaray Alvarez (Banco Central del Uruguay) |
| Abstract: | This paper estimates, evaluates and compares the nowcasting and forecast¬ing performance of ADL-MIDAS, U-MIDAS and TF-MIDAS models. To do so, 25 monthly series are considered as possible indicators of GDP growth and the forecast accuracy is assessed in two overlapping periods: one period that includes the COVID-19 pandemic and another that excludes it. When the period affected by the pandemic is considered TFMIDAS results the most precise model, while if this period is excluded from the evaluation there is no difference in the predictive ability between the three models considered. Then different forms of nonparametric forecasts combinations are evaluated, obtaining similar results with most of them and, in general, outperforming when the best model is the only one considered. |
| Keywords: | Macroeconomic forecasting, Mixed frequency, TF-MIDAS, U-MIDAS, ADL-MIDAS, Nowcasting, Forecasts combination |
| JEL: | C22 C32 C53 E37 |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:bku:doctra:2022004&r=for |
| By: | Alexander Chudik; M. Hashem Pesaran; Mahrad Sharifvaghefi |
| Abstract: | This paper is concerned with the problem of variable selection when the marginal effects of signals on the target variable as well as the correlation of the covariates in the active set are allowed to vary over time, without committing to any particular model of parameter instabilities. It poses the issue of whether weighted or unweighted observations should be used at the variable selection stage in the presence of parameter instability, particularly when the number of potential covariates is large. Amongst the extant variable selection approaches, we focus on the One Covariate at a time Multiple Testing (OCMT) method. This procedure allows a natural distinction between the selection and forecasting stages. We establish three main theorems on selection, estimation post selection, and in-sample fit. These theorems provide justification for using unweighted observations at the selection stage of OCMT and down-weighting of observations only at the forecasting stage. The benefits of the proposed method as compared to Lasso, Adaptive Lasso and Boosting are illustrated by Monte Carlo studies and empirical applications to forecasting monthly stock market returns and quarterly output growths. |
| Keywords: | parameter instability, high-dimensionality, variable selection, One Covariate at a time Multiple Testing (OCMT) |
| JEL: | C22 C52 C53 C55 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ceswps:_10223&r=for |