Mixed Data Sampling Model Research Articles

Density Forecasts with MIDAS Models

In this paper we propose a parametric block wild bootstrap approach to compute density forecasts for various types of mixed-data sampling (MIDAS) regressions. First, Monte Carlo simulations show that predictive densities for the various MIDAS models derived from the block wild bootstrap approach are more accurate, in terms of coverage rates, than predictive densities derived from either a residual-based bootstrap approach or by drawing errors from a normal distribution. This result holds irrespective of the data generating errors being normally independently distributed, serially correlated, heteroskedastic or a mixture of normal distributions. Second, in an empirical exercise we evaluate density forecasts for quarterly U.S. real output growth, exploiting information from typical monthly and weekly series. We show that the block wild bootstrapping approach, applied to the various MIDAS regressions, produces predictive densities for U.S. real output growth that are well-calibrated. Moreover, the relative accuracy, measured in terms of the logarithmic score or the continuous ranked probability score, improves for the various MIDAS specifications as more information becomes available.

A MIDAS modelling framework for Chinese inflation index forecast incorporating Google search data

Increased internet penetration makes it possible for user generated content (UGC) to reflect people’s insights and expectations on economic activities. As representative and easily accessible UGC data that reflect public opinions on economic issues, Google search data have been used to forecast macroeconomic indicators in existing literatures. However, very little empirical research has directly used Google search data to improve the forecast accuracy. This paper proposes an integrated framework, which constructs keywords base and extracts search data accordingly, and then incorporates the search data into a mixed data sampling (MIDAS) model. Five groups of search data are extracted based on the constructed keywords and are then used in MIDAS model to forecast Chinese consumer price index (CPI) from 2004 to 2012. The empirical results indicate that the search data are strongly correlated with CPI, which is officially released by the Statistic Bureau of China; the MIDAS model including the search data outperforms the benchmark models, with the average reduction of root mean square error (RMSE) being 32.9%. This research provides a rigorous and generalizable framework for macroeconomic trend prediction using Google search data, and would have great potential in supporting business decisions by eliciting relevant information from UGC data in the Internet.

Variable Selection in Predictive MIDAS Models

In short-term forecasting, it is essential to take into account all available information on the current state of the economic activity. Yet, the fact that various time series are sampled at different frequencies prevents an efficient use of available data. In this respect, the Mixed-Data Sampling (MIDAS) model has proved to outperform existing tools by combining data series of different frequencies. However, major issues remain regarding the choice of explanatory variables. The paper first addresses this point by developing MIDAS based dimension reduction techniques and by introducing two novel approaches based on either a method of penalized variable selection or Bayesian stochastic search variable selection. These features integrate a cross-validation procedure that allows automatic in-sample selection based on recent forecasting performances. Then the developed techniques are assessed with regards to their forecasting power of US economic growth during the period 2000-2013 using jointly daily and monthly data. Our model succeeds in identifying leading indicators and constructing an objective variable selection with broad applicability.

Forecasting GDP growth using mixed-frequency models with switching regimes

For modelling mixed-frequency data with a business cycle pattern, we introduce the Markov-switching Mixed Data Sampling model with unrestricted lag polynomial (MS-U-MIDAS). Usually, models of the MIDAS-class use lag polynomials of a specific function which impose some structure on the weights of the regressors included in the model. This may lead to a deterioration in the predictive power of the model if the structure imposed differs from the data generating process. When the difference between the available data frequencies is small and there is no risk of parameter proliferation, using an unrestricted lag polynomial might not only simplify the model estimation, but also improve its forecasting performance. We allow the parameters of the MIDAS model with an unrestricted lag polynomial to change according to a Markov-switching scheme in order to account for the business cycle pattern observed in many macroeconomic variables. Thus, we combine the unrestricted MIDAS with a Markov-switching approach and propose a new Markov-switching MIDAS model with unrestricted lag polynomial (MS-U-MIDAS). We apply this model to a large dataset with the help of factor analysis. Monte Carlo experiments and an empirical forecasting comparison carried out for the U.S. GDP growth show that the models of the MS-U-MIDAS class exhibit nowcasting and forecasting performances which are similar to or better than those of their counterparts with restricted lag polynomials.

Nowcasting and Short-term Forecasting of Chinese Quarterly GDP: Mixed Frequency Approach

Conventional macroeconomic forecasting model must change the mixed frequency data into the same frequency data by means of aggregation and interpolation, which ignores the information of high frequency data and decreases the timeliness and accuracy of nowcasting and short-term forecasting. Using monthly and quarterly macroeconomic data, this paper apply mixed data sampling (MIDAS) model and mixed-frequency VAR model (MF-VAR) to nowcast and forecast Chinese quarterly GDP growth rate. The results show that MIDAS model considering an autoregressive item tends to perform better in shorter horizons, whereas MF-VAR model in longer horizons. Additionally, the pooled forecast of MIDAS model and MF-VAR model outperform the individual models. Finally, forecasting results reveal that China's quarterly GDP growth will rebound steadily from the beginning of the third quarter of 2012

MIDAS volatility forecast performance under market stress: Evidence from emerging stock markets

This paper evaluates weekly out-of-sample volatility forecast performance of univariate Mixed Data Sampling (MIDAS) model compared to the benchmark model of GARCH(1,1) for ten emerging stock markets. The results show that the MIDAS model offers a statistically better forecasting precision during the recent financially turbulent era, based on the test suggested by West (2006). For the tranquil period, however, the MIDAS model cannot produce a statistically better weekly volatility forecast.

Predicting Volatility: Getting the Most out of Return Data Sampled at Different Frequencies

We use the MIDAS (Mixed Data Sampling) approach to study regressions of future realized volatility at low-frequency horizons (one to four weeks) on lagged daily and intra-daily (1) squared returns, (2) absolute returns, (3) realized volatility, (4) realized power and (5) return ranges. We document first of all that daily realized power and daily range are surprisingly good predictors of future realized volatility and outperform models based on realized volatility. Moreover, MIDAS models with daily data - range, realized power, realized volatility - require a polynomial with at least 30 days. We document that high-frequency absolute returns are also better at forecasting future low frequency realized volatility than high-frequency squared returns. We also discuss many issues that are encountered in practice, such as long memory and seasonality. All the results are based on a commonly used FX data set.