MIDAS regressions: Further results and new directions
read more
Citations
Machine learning panel data regressions with heavy-tailed dependent data: Theory and application
Sample frequency robustness and accuracy in forecasting Value-at-Risk for Brent Crude Oil futures
Improving model-based near-term GDP forecasts by subjective forecasts: A real-time exercise for the G7 countries
The R package sentometrics to compute, aggregate and predict with textual sentiment
Nowcasting Chinese GDP in a data-rich environment: Lessons from machine learning algorithms
References
Investigating Causal Relations by Econometric Models and Cross-Spectral Methods
Investigating causal relations by econometric models and cross-spectral methods
Conditional heteroskedasticity in asset returns: a new approach
On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks
An intertemporal capital asset pricing model
Related Papers (5)
Predicting volatility: getting the most out of return data sampled at different frequencies
Frequently Asked Questions (7)
Q2. How many lags would be needed to estimate yt?
For instance, if monthly observations of yt is affected by six months’ worth of lagged daily x (m) t ’s, the authors would need 132 lags (K = 132) of high-frequency lagged variables.
Q3. Why do the authors report the mean absolute deviation (MAD)?
The authors report the mean absolute deviation (MAD) as a measure of goodness of fit (fourth column), because it provides more robust results in the presence of heteroskedasticity.
Q4. What is the function B(L1/m; ) of a few parameters?
As a way of addressing parameter proliferation, in a MIDAS regression the coefficients of the polynomial in L1/m are captured by a known function B(L1/m; θ) of a few parameters summarized in a vector θ.
Q5. What is the predictor of conditional volatility?
Santa-Clara, and Valkanov (2003) show that the best overall predictor of conditional volatility is the realized power and that, not surprisingly, better forecasts are obtained at shorter (weekly) horizons.
Q6. How many parameters are there to estimate?
If the parameters of the lagged polynomial are left unrestricted (or B(k) does not depend on θ), then there would be a lot of parameters to estimate.
Q7. What is the way to model the volatility of financial markets?
It is also worth noting that for stochastic volatility models the problem is even more difficult since the volatility factors are latent and therefore need to be extracted from observed past returns.