Nowcasting as the new potential predicting method for the policy-makers

• Thesis title in Czech: Nowcasting jako nová prediktivní metoda pro rozhodování politiků
• Thesis title in English: Nowcasting as the new potential predicting method for the policy-makers
• Key words: Nowcasting, předpovídání, přesnost, HDP, rozhodování
• English key words: Nowcasting, Forecasting, Accuracy, GDP, Policy-making
• Academic year of topic announcement: 2021/2022
• Thesis type: Bachelor's thesis
• Thesis language: angličtina
• Department: Department of Political Science (23-KP)
• Supervisor: RNDr. Jan Kofroň, Ph.D.
• Author: hidden - assigned by the advisor
• Date of registration: 04.05.2022
• Date of assignment: 04.05.2022
• Date and time of defence: 11.09.2023 08:00
• Venue of defence: Jinonice - Nový Kampus, B328, 328, seminární místnost IMS
• Date of electronic submission: 30.07.2023
• Date of proceeded defence: 11.09.2023
• Opponents: Dr. rer. pol. Michal Parízek, M.Sc., Ph.D.

References

STÄHLER, Nikolai, Carlos THOMAS. FiMod — A DSGE model for fiscal policy simulations. Economic modelling [online]. Amsterdam: Elsevier B.V, 2012, 29(2), 239-261 [cit. 2022-08-06]. ISSN 0264-9993. Dostupné z: doi:10.1016/j.econmod.2011.10.001

ADAM, Tomas, Ondrej MICHALEK, Ales MICHL a Eva SLEZAKOVA. The Rushin Index: A Weekly Indicator of Czech Economic Activity. IDEAS Working Paper Series from RePEc [online]. St. Louis: Federal Reserve Bank of St Louis, 2021.
Keqiang ker-ching; China's economy. The Economist (London) [online]. Economist Intelligence Unit N.A. Incorporated, 2010, 397(8712), 54 [cit. 2022-08-06]. ISSN 0013-0613
CHEN, Sophia. Tracking the Economic Impact of COVID-19 and Mitigation Policies in Europe and the United States. IDEAS Working Paper Series from RePEc [online]. St. Louis: Federal Reserve Bank of St Louis, 2020 [cit. 2022-08-06].

DIEBOLD, Francis X., Roberto S. MARIANO. Comparing Predictive Accuracy. Journal of business & economic statistics [online]. Washington, D.C: Taylor & Francis Group, 1995, 13(3), 253-263 [cit. 2022-08-06]. ISSN 0735-0015. Dostupné z: doi:10.1080/07350015.1995.10524599

MINCER, Jacob A., Victor ZARNOWITZ (1969): The Evaluation of Economic Forecasts, Economic Forecasts and Expectations: Analysis of Forecasting Behavior and Performance
ADAM, Tomáš, Aleš MICHL (2020). “První odhad dopadů pandemie COVID-19 na ekonomiku ČR”. čnBlog – oficiální blog České národní banky, https://www.cnb.cz/cs/o_cnb/cnblog/Prvni-odhad-dopadu-pandemie-COVID-19-na-ekonomiku-CR/ .

ERASLAN, Sercan, Thomas GÖTZ (2021). “An Unconventional Weekly Economic Activity Index For Germany”. Economics Letters, 204:109881

GASCON, Charles S., Amelia SCHMITZ (2020). “Using High-Frequency Data to Track the Regional Economy”. The Regional Economist. 28(3)

WOLOSZKO, Nicolas (2020), "Tracking activity in real time with Google Trends", OECD Economics Department Working Papers, No. 1634, OECD Publishing, Paris, https://doi.org/10.1787/6b9c7518-en.

Preliminary scope of work in English

Research question and motivation
Predictions are one of the basic things in every science, in every field, including political science and economy. The accuracy of predictions depends on many variables whose differences can sometimes result in very different outcomes. These outcomes of the predictions are mostly used in simple decisions, such as whether the weather will be suitable for a walk today, to very key decisions at the highest levels concerning military operations or macroeconomic decisions that will affect economic developments in the coming months and years. We distinguish between three basic types of prediction models, forecast, which is the predict of the future, nowcast, which predicts the current state and close future, and backcast, predicting what has already happened.
Since macroeconomic forecast model named Dynamic Stochastic General Equilibrium (DSGE), which is widely used, is made quarterly based on robust product level data, the performance is timely limited, especially in dynamic times, because the model assumes linearity of state intervention (Stähler, 2012). In contrast, nowcasting uses high-frequency data that are published daily or even hourly, for example Google trends or toll collection on the roads, to predict the current state (Adam, 2021). Yet nowcasting is a new tool for decision makers, for example, Keqiang index for China’s economy compiled by The Economist in 2010, was followed in 2020 (Chen) tracking the economic impact of COVID-19.
As COVID-19 pandemic came, as the exogenous shock, the state invention became more unpredictable and sudden as the virus spread, which means that forecasting models have become less useful, even unusable. Therefore, my main research questions for my bachelor thesis are: How the forecast models perform during recession and may be the nowcasting indexes be more useful for the decision makers?

Contribution
Firstly, I will evaluate the forecast indexes of GDP, especially during the endogenous and exogenous economic recessions, with a goal to answer the question of whether economic forecast may be useful during recession? Then I will introduce nowcasting indexes, evaluate their performance and explain why and whether they might be more useful for decision makers than forecast.

Methodology
I will evaluate the performance of forecasts in times of endogenous and exogenous recessions using the combination of Diebold-Mariano test, which tests whether the difference of two forecasts is significant (Diebold, 1995), and Mincer-Zaronowitz regression, which identify systemic errors (Mincer, 1969), using dataset from IMF, OECD and EC. I will introduce nowcasting indexes constructed by Eraslan and Götz (2021), of weekly activity index for Germany, Gascon and Schmitz (2020), monitoring regional development in the US, Woloszko (2020), tracking economic activity in OECD, Adam and Michl (2020), tracking economy activity of the Czech Republic, etc. I will evaluate their performance and explain how the newly used nowcasting could benefit decision makers.