Browsing by Subject "E27"

Sort by: Order: Results:

Now showing items 1-4 of 4
  • Bailliu, Jeannine; Han, Xinfen; Kruger, Mark; Liu, Yu-Hsien; Thanabalasingam, Sri (2018)
    BOFIT Discussion Papers 9/2018
    Published in International Journal of Forecasting, 35, 3, 2019, 1118-1130
    The official Chinese labour market indicators have been seen as problematic, given their small cyclical movement and their only-partial capture of the labour force. In our paper, we build a monthly Chinese labour market conditions index (LMCI) using text analytics applied to mainland Chinese-language newspapers over the period from 2003 to 2017. We use a supervised machine learning approach by training a support vector machine classification model. The information content and the forecast ability of our LMCI are tested against official labour market activity measures in wage and credit growth estimations. Surprisingly, one of our findings is that the much-maligned official labour market indicators do contain information. However, their information content is not robust and, in many cases, our LMCI can provide forecasts that are significantly superior. Moreover, regional disaggregation of the LMCI illustrates that labour conditions in the export-oriented coastal region are sensitive to export growth, while those in inland regions are not. This suggests that text analytics can, indeed, be used to extract useful labour market information from Chinese newspaper articles.
  • Mikosch, Heiner; Solanko, Laura (2019)
    Russian Journal of Money and Finance 1
    This paper presents a pseudo real‐time out‐of‐sample forecast exercise for short‐term forecasting and nowcasting quarterly Russian GDP growth with mixed‐frequency data. We employ a large set of indicators and study their predictive power for different subperiods within the forecast evaluation period 2008–2016. Four indicators consistently figure in the list of top-performing indicators: the Rosstat key sector economic output index, the OECD composite leading indicator for Russia, household banking deposits, and money supply M2. Aside from these indicators, the top indicators in the 2008–2011 evaluation period are traditional real‐sector variables, while those in the 2012–2016 evaluation period largely comprise monetary, banking sector and financial market variables. We also compare the forecast accuracy of three different mixed‐frequency forecasting model classes (bridge equations, MIDAS models, and U-MIDAS models). Differences between the performance of model classes are generally small, but for the 2008–2011 period MIDAS models and U-MIDAS models outperform bridge equation models.
  • Mikosch, Heiner; Neuwirth, Stefan (2015)
    BOFIT Discussion Papers 13/2015
    This paper presents a MIDAS type mixed frequency VAR forecasting model. First, we propose a general and compact mixed frequency VAR framework using a stacked vector approach. Second, we integrate the mixed frequency VAR with a MIDAS type Almon lag polynomial scheme which is designed to reduce the parameter space while keeping models fexible. We show how to recast the resulting non-linear MIDAS type mixed frequency VAR into a linear equation system that can be easily estimated. A pseudo out-of-sample forecasting exercise with US real-time data yields that the mixed frequency VAR substantially improves predictive accuracy upon a standard VAR for dierent VAR specications. Forecast errors for, e.g., GDP growth decrease by 30 to 60 percent for forecast horizons up to six months and by around 20 percent for a forecast horizon of one year.
  • Mikosch, Heiner; Solanko, Laura (2017)
    BOFIT Discussion Papers 19/2017
    Published in Russian Journal of Money and Finance, Volume 78, Number 1, March 2019
    GDP forecasters face tough choices over which leading indicators to follow and which forecasting models to use. To help resolve these issues, we examine a range of monthly indicators to forecast quarterly GDP growth in a major emerging economy, Russia. Numerous useful indicators are identified and forecast pooling of three model classes (bridge models, MIDAS models and unrestricted mixed-frequency models) are shown to outperform simple benchmark models. We further separately examine forecast accuracy of each of the three model classes. Our results show that differences in performance of model classes are generally small, but for the period covering the Great Recession unrestricted mixed-frequency models and MIDAS models clearly outperform bridge models. Notably, the sets of top-performing indicators differ for our two subsample observation periods (2008Q1–2011Q4 and 2012Q1–2016Q4). The best indicators in the first period are traditional real-sector variables, while those in the second period consist largely of monetary, banking sector and financial market variables. This finding supports the notion that highly volatile periods of recession and subsequent recovery are driven by forces other than those that prevail in more normal times. The results further suggest that the driving forces of the Russian economy have changed since the global financial crisis.