|
|
Nowcasting as the new potential predicting method for the policy-makers
Chaloupka, David ; Kofroň, Jan (advisor) ; Parízek, Michal (referee)
This bachelor's thesis examines the role of nowcasting as a real-time data pre- diction approach in policy-making. The study compares the informative value of macroeconomic forecasting during crises and the potential of nowcasting as an alternative. By analyzing selected forecasting and nowcasting indexes from 2019 to 2022, the research fnds that nowcasting performs well during during 2020, which was afected by COVID-19 pandemic. However, it faces limitations in terms of prediction horizon and data requirements. Although forecasting may lose its informative value during crises, nowcasting cannot entirely replace it. Instead, both approaches can complement each other, enhancing policy decisions. The thesis also highlights nowcasting's potential for policy analysis and also its use in science.
|
|
|
Nowcasting the Real GDP Growth of the European Economies based on Machine Learning
Baylan, Su Hazal ; Kočenda, Evžen (advisor) ; Baruník, Jozef (referee)
This thesis analyzes the nowcasting of quarterly GDP growth for nine European economies using a dynamic factor model and four different machine learning models. These machine learning models are as follows: Ridge, Lasso, Elastic Net, and Random Forest. The data includes ten hard and fifteen soft indicators for each country in order to calculate GDP for each nowcasting iteration for pre-covid and covid periods. For machine learning, models are fed with the extracted factors that are obtained from the dynamic factor model, and for all nowcasting models expanding window approach is selected to estimate nowcasting iterations. The empirical finding indicates that overall machine learning models provide better forecasting accuracy compared to dynamic factor models and benchmark models for more stable periods, such as the period before Covid-19. On the other hand, for more volatile periods where the uncertainties are higher in economies, the dynamic factor model outperforms machine learning models in order to nowcast GDP growth. In addition to this, Random Forest is able to outperform all the alternative models for small economies such as Slovenia and Portugal for stable periods. JEL Classification C01, C33, C53, C83, E37 Keywords Nowcasting, DFM, Ridge, Lasso, Elastic Net, Random Forest Title Nowcasting...
|
|
| |
|
|
Forecasting and nowcasting power of confidence indikators:Evidence for Central Europe
Herrmannová, Lenka ; Horváth, Roman (advisor) ; Mikolášek, Jakub (referee)
This thesis assesses the usefulness of confidence indicators for nowcasting and short term forecasting of the economic activity in the Czech Republic and three other Central European countries. The predictive power of both the Czech business confidence indicator and the customer confidence indicator is examined using two empirical approaches. First we predict the likelihood of economic downturn using logit models, later we estimate GDP growth out of sample forecasts in the framework of vector autoregression models. The results obtained from the downturn probability models confirm the ability of confidence indicators (especially the business confidence indicator) to estimate the current economic situation, so called nowcast. Results from the out-of-sample GDP growth value forecasting are ambiguous. Nevertheless the customer confidence indicator significantly improved original forecasts based on the model with standard macroeconomic variables and therefore we conclude in favour of its predictive power. Cross- country comparison confirms economic downturn nowcasting power of confidence indices in Hungary and Poland and fails to confirm such an ability of Slovak confidence indicators. One-quarter-ahead forecasts brought mixed results and therefore we conclude that nowcasting and forecasting properties of...
|
|
|
Precipitation nowcasting for the warm part of the year
Mejsnar, Jan ; Sokol, Zbyněk (advisor) ; Jaňour, Zbyněk (referee) ; Žák, Michal (referee)
Current precipitation nowcasting systems primarily use the extrapolation of observed radar reflectivity. I used the extrapolation and studied limits of the forecast using the concept of the decorrelation time (DCT). I used data from two radars covering the territory of the Czech Republic from warm parts of four years and calculated DCT in dependence on several selected conditions describing the state of the atmosphere. I found that the mean DCT for the extrapolation is 45.4 minutes. On average the increase of the DCT in comparison when the persistence forecast is employed is 13.4 minutes. However, in dependence on current conditions the DCT may increase or decrease in more than 40 %. I also explored time evolution of the DCT during two storm events. I found that the DCT may significantly change in time, which is the consequence of changing character of the atmosphere during the storm development.
|
|
|
Utilization of Radar Echo Extrapolation for Quantitative Precipitation Forecast
Frolík, Petr ; Novák, Petr (advisor) ; Žák, Michal (referee)
Title: Utilization of radar echo extrapolation for quantitative precipitation forecast Author: Petr Frolík Department: Department of Meteorology and Enviroment Protection Supervisor: RNDr. Petr Novák, PhD., ČHMÚ Supervisor's e-mail address: petr_novak@chmi.cz Abstract: At present time weather radar data are essential for national meteorological services. Utilization of this data for quantitative precipitation forecast and severe weather prediction for short period (nowcasting) becomes more and more common. Increasing interest for quantitative precipitation forecasts can be noticed in hydrological applications, where it can give early warning on flash floods and can improve large scale precipitation forecasts. This paper verifies usability of COTREC nowcasting method based on extrapolation of radar echo for quantitative precipitation forecast. Quality of COTREC forecast up to 3 hours was investigated on data from 1.4.2006 to 30.9.2006. Comparison of COTREC method with Aladin NWP model forecasts was also made. Hourly mean precipitation estimates on catchments were chosen for comparison because of verification focused mainly on utilization of the forecasts in hydrological applications. Forecasted precipitation estimates were compared with optimal operationally available precipitation estimate - adjusted radar...
|
|
| |
|
|
Forecasting oil prices volatility with Google searches
Tolstoguzova, Ekaterina ; Krištoufek, Ladislav (advisor) ; Zafeiris, Dimitrios (referee)
Oil market pricing is highly susceptible to geopolitical and economic events. With the rapid development of information technology, energy market can quickly get external information shocks through the Internet. This thesis examines the relationship between prices of three oil benchmarks, CBOE Crude Oil Volatility Index, and Google search queries. We built VAR model to study Granger causality and to provide impulse response analysis. Results indicate both one side and two-side causal relationship between oil-related series and most of the search queries. Out-of sample forecasting with measures of predictive accuracy and Diebold-Mariano test demonstrated that Google trends can improve short-run prediction potential only for models with WTI price and volatility index.
|
|
|
Precipitation nowcasting for the warm part of the year
Mejsnar, Jan ; Sokol, Zbyněk (advisor) ; Jaňour, Zbyněk (referee) ; Žák, Michal (referee)
Current precipitation nowcasting systems primarily use the extrapolation of observed radar reflectivity. I used the extrapolation and studied limits of the forecast using the concept of the decorrelation time (DCT). I used data from two radars covering the territory of the Czech Republic from warm parts of four years and calculated DCT in dependence on several selected conditions describing the state of the atmosphere. I found that the mean DCT for the extrapolation is 45.4 minutes. On average the increase of the DCT in comparison when the persistence forecast is employed is 13.4 minutes. However, in dependence on current conditions the DCT may increase or decrease in more than 40 %. I also explored time evolution of the DCT during two storm events. I found that the DCT may significantly change in time, which is the consequence of changing character of the atmosphere during the storm development.
|
|
| |
guest ::
