MIDAS Regressions: Further Results and New Directions

By Eric Ghysels, A. Sinko, Rossen Valkanov. 2007

1,019 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/4d02f73c1acf1af08090639f166ccde8f691e2bc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Finance

Citations in the corpus

A MIDAS Approach to Modeling First and Second Moment Dynamics (2016, neutral): “A third alternative is the stepwise weights proposed in Ghysels et al. (2007) and Corsi (2009) B (k; θ) = θ 1 I k∈[a 0 ,a 1 ] + p p=2 θ p I k∈[a p−1 ,ap] , where the p + 1 parameters a 0 , ..., a p are thresholds for the step function with a 0 = 1 < a 1 < ... < a p = K and I k∈[a p−1 ,ap] is an…”
Exogenous drivers of Bitcoin and Cryptocurrency volatility – A mixed data sampling approach to forecasting (2019, methodology): “To weight the lagged low-frequency variables, we use the Beta-weighting scheme φk (ω1, ω2) introduced by Ghysels et al. (2007). It is straightforward that the model decomposes to the standard GARCH model if θ = 0 in Eq.”
Harnessing jump component for crude oil volatility forecasting in the presence of extreme shocks (2019, methodology): “We follow the convention to set k max =403, and follow Ghysels et al. (2002, 2006, 2007), Ghysels and Valkanov (2012), and Forsberg and Ghysels (2007) to set up the weight function kw using the following Beta function: 3 The value of 40 is determined by grid search.”
Out of sample forecasts of quadratic variation (2008, background): “Further results are reported in Ghysels and Sinko (2006b) and Ghysels et al. (2008) , whose data generating process for the volatility is of the same type as in the papers just cited, but explicitly include market microstructure noise.”
Macroeconomics and the Reality of Mixed Frequency Data (2016, methodology): “There are various parsimonious parameterizations suggested for such regressions, see e.g. Ghysels et al. (2006), Andreou et al. (2010) and Sinko et al. (2010), which will be discussed later.”
HAR Modeling for Realized Volatility Forecasting (2012, neutral): “By substitution, this gives: where ε t is iid noise depending on ε In this sense, the HAR can be related to the MIDAS regression of [47], [48], and [45], although the standard MIDAS with the estimated Beta function lag polynomial cannot reproduce the HAR step function weights.”
Forecasting with Mixed Frequencies (2010, background): “Ghysels, Santa-Clara, and Valkanov (2005) and Ghysels, Sinko, and Valkanov (2007) also suggest an exponential Almon specification: In this case, simple time averaging is obtained when θ 1 = θ 2 = 0.”
Testing for Granger Causality in Large Mixed-Frequency VARs (2016, neutral): “As argued extensively in the mixed-frequency literature (e.g., Ghysels et al., 2007), working in a mixed-frequency setup instead of a common low-frequency one is advantageous due to the potential loss of information in the latter scenario and feasibility of the former through MI(xed) DA(ta)…”
Predicting the long-term stock market volatility: A GARCH-MIDAS model with variable selection (2020, methodology): “their “Beta weighting” schemes of the MIDAS model (Ghysels et al., 2007).”
Multi-Period Forecasts of Volatility: Direct, Iterated, and Mixed-Data Approaches (2009, background): “First, we could have extended the number of regression to include not only daily squared returns but other volatility forecasts such as daily absolute returns, daily range measures (high-low), and others, as done in Ghysels et al. (2006).”