Using low frequency information for predicting high frequency variables

Foroni, Claudia; Guérin, Pierre; Marcellino, Massimiliano

doi:10.1016/j.ijforecast.2018.06.004

Cited by 54 publications

(35 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Empirically, we find no major differences between forecasts based either on MF-3PRF or on 3PRF with interpolated monthly time series; see Section 3.4. To account for a high frequency dependent variable and a low frequency regressor, one could also use the reverse MIDAS or reverse U-MIDAS approaches of Foroni, Guerin and Marcellino (2015). However, as the model in pass 1 is static, no major gains are expected in our context.…”

Section: Low Frequency Target or Proxy Variables And High Frequency Imentioning

confidence: 99%

Forecasting Gross Domestic Product Growth with large Unbalanced Data Sets: The Mixed Frequency Three-pass Regression Filter

Hepenstrick

Marcellino

2018

Journal of the Royal Statistical Society Series A: Statistics in Society

Self Cite

View full text Add to dashboard Cite

Summary Gross domestic product (GDP) is a key summary of macroeconomic conditions and it is closely monitored both by policy makers and by decision makers in the private sector. However, it is only available on a quarterly frequency, and in many countries it is released with a substantial delay. There are, however, many higher frequency and more timely economic and financial indicators that could be used for nowcasting and short‐term forecasting GDP. Against this backdrop, we propose a modification of the three‐pass regression filter to make it applicable to large mixed frequency data sets with ragged edges in a forecasting context. The resulting method, labelled MF‐3PRF, is very simple but compares well with alternative mixed frequency factor estimation procedures in terms of theoretical properties and actual GDP nowcasting and forecasting for the USA and a variety of other countries.

show abstract

Section: Low Frequency Target or Proxy Variables And High Frequency Imentioning

confidence: 99%

Forecasting Gross Domestic Product Growth with large Unbalanced Data Sets: The Mixed Frequency Three-pass Regression Filter

Hepenstrick

Marcellino

2018

Journal of the Royal Statistical Society Series A: Statistics in Society

Self Cite

View full text Add to dashboard Cite

show abstract

“…At the same time, daily data-based HARtype models too, where realized volatility (RV) estimates are derived from intraday data, do not allow for predictors at lower frequency, as would be the case in our context. This in turn would require the usage of reverse-MIDAS regressions recently developed by Foroni et al (2018), which could indeed be an area of future research. Of course, we could obtain RV at monthly frequency from daily data and use a monthly version of the HAR-RV model, but then this would not allow us to produce daily forecasts.…”

Section: Introductionmentioning

confidence: 99%

Mixed‐frequency forecasting of crude oil volatility based on the information content of global economic conditions

Salisu

Gupta

Bouri

et al. 2021

Journal of Forecasting

View full text Add to dashboard Cite

show abstract

“…16 Third, the forecasts are constructed as 15 This result still holds when doing PCA at a monthly frequency and then aggregating the factor at a quarterly frequency. 16 As a side note, the first pass of the 3PRF filter could possibly accommodate mixed-frequency data using the techniques outlined in Foroni et al (2015); whereas, in the third pass of the filter, unrestricted mixed data sampling (MIDAS) polynomials could be used as in Hepenstrick and Marcellino (2016), and regime-switching parameters in the mixed-frequency predictive equation could be modelled as in Guérin and Marcellino (2013).…”

Section: Forecasting Economic Activitymentioning

confidence: 99%

Markov-Switching Three-Pass Regression Filter

Guérin

Leiva-León

Marcellino

2018

Journal of Business & Economic Statistics

Self Cite

View full text Add to dashboard Cite

We introduce a new approach for the estimation of high-dimensional factor models with regime-switching factor loadings by extending the linear three-pass regression filter to settings where parameters can vary according to Markov processes. The new method, denoted as Markov-switching three-pass regression filter (MS-3PRF), is suitable for datasets with large cross-sectional dimensions, since estimation and inference are straightforward, as opposed to existing regime-switching factor models where computational complexity limits applicability to few variables. In a Monte Carlo experiment, we study the finite sample properties of the MS-3PRF and find that it performs favourably compared with alternative modelling approaches whenever there is structural instability in factor loadings. For empirical applications, we consider forecasting economic activity and bilateral exchange rates, finding that the MS-3PRF approach is competitive in both cases. Bank topic: Econometric and statistical methods JEL codes: C22; C23; C53 RésuméNous introduisons une nouvelle méthode d'estimation des modèles factoriels de large dimension avec changements de régimes en élargissant le filtre de régression linéaire à trois passages afin d'intégrer des paramètres qui peuvent varier en fonction de processus de Markov. Cette nouvelle méthode, appelée filtre de régression à trois passages avec changements de régimes markoviens (MS-3PRF), convient à des ensembles de données dotés d'importantes dimensions transversales car l'estimation et l'inférence s'y font directement. À l'inverse, les modèles factoriels à changements de régimes existants ne peuvent s'employer qu'avec un petit nombre de variables compte tenu de la complexité des algorithmes utilisés pour estimer ces modèles. Nous nous servons d'une expérience de Monte Carlo pour étudier les propriétés du filtre MS-3PRF sur un échantillon fini. Ce filtre donne de bons résultats par rapport aux autres méthodes de modélisation lorsque les pondérations factorielles présentent une instabilité structurelle. Pour l'application empirique, nous examinons la qualité des prévisions de l'activité économique et des taux de change bilatéraux : dans les deux cas, le filtre MS-3PRF se compare favorablement aux autres approches. Sujet : Méthodes économétriques et statistiques Codes JEL : C22 ; C23 ; C53Non-Technical Summary This paper introduces a new approach for the estimation of high-dimensional factor models with regime-switching factor loadings. The premise of factor models is that the common dynamics of a large number of variables can be summarized in a relatively small number of unobserved factors. The literature on factor models has mostly concentrated on situations where the comovement among variables is assumed to be constant over time. However, there is now a large body of literature that has challenged the assumption of constant parameters to model the macroeconomic environment. It is also often found important to model time variation for macroeconomic forecasting.The first contribution of this paper...

show abstract

Using low frequency information for predicting high frequency variables

Cited by 54 publications

References 42 publications

Forecasting Gross Domestic Product Growth with large Unbalanced Data Sets: The Mixed Frequency Three-pass Regression Filter

Forecasting Gross Domestic Product Growth with large Unbalanced Data Sets: The Mixed Frequency Three-pass Regression Filter

Mixed‐frequency forecasting of crude oil volatility based on the information content of global economic conditions

Markov-Switching Three-Pass Regression Filter

Contact Info

Product

Resources

About