Search

Exogenous shocks like the COVID-19 pandemic or the Russian conflict affected the global economy, but mostly emerging economies. Even though the critical moments have passed, the negative effects on the economy are still present in the current economic environment, characterized by high levels of uncertainty. Exogenous events, which have an uncertain likelihood of occurring, harm economies regardless of their development. In fact, a framework for unanticipated events is vital, as are steps to achieve low levels of national debt and increased authority stability. In this context, the objective of this paper is to examine the dynamic relationships between energy prices and certain macroeconomic variables, which were affected by the fact that the energy crisis triggered by the Russian-Ukrainian conflict overlapped with the accelerated transition to a green economy and the decarbonization process. The CEE countries are more vulnerable to the energy crisis, and their proximity to the front lines of military conflict can be an important fact to take into account. The research focuses on the Czech Republic, Hungary, Poland, and Romania. The empirical analysis was conducted on monthly frequency data and in order to present an effective method of modeling interactions between macroeconomic variables, the Bayesian technique was employed.

The aim of this study is analyzing the spillovers of monetary policies from three nations including the US, Japan, and China to the Southeast Asian economies through macroeconomic linkages. By using BVAR model, the research results present distinct responses of economic growth, interest rate and inflation index in Southeast Asia to external shocks of interest rate according to specific characteristics. In particular, nations adopting pegged regime have stronger responses to the changing of the Fed rate and Japanese monetary policy while the opposite trend is found in volatility of Chinese interest rate. However, nations with higher trade openness might have more sensitive reaction to monetary policy of all three countries. On the other hand, the impact of international monetary policies on Southeast Asian countries is explained quite different between the group having higher capital openness and the other one. In general, the results are consistent with many previous studies as well as Mundell-Fleming's impossibility triple theory. [ABSTRACT FROM AUTHOR]

The Greening in agro—food sector has become within the last decade a high priority issue given the 17 Sustainable targets set by OECD. More specifically, the Sustainable Development Goals (SDGs) by 2030, intend to promote using environmental resources in close correlation with measures to reduce non-environmental human pressure on the planet as well as in agro-food sector. The present work studies the greening of agro-food sector as synopsized in emissions per capita by agro-food sector for the EU and its relation to economic growth per capita with the assistance of a BVAR framework. Our findings do not validate success in greening of agro—food sector since the emissions reduction is not accompanied by economic growth a result that rejects the hypothesis of eco efficiency. Future research could involve the construction of an index that should incorporate more variables that will reflect more accurately the greening efforts in agro—food sector. [ABSTRACT FROM AUTHOR]

Copyright of GeSec: Revista de Gestao e Secretariado is the property of Sindicato das Secretarias e Secretarios do Estado de Sao Paulo (SINSESP) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

We apply a Markov-switching Bayesian vector autoregression (MSBVAR) model to mortality forecasting. MSBVAR has not previously been applied in this context, and our results show that it is a promising tool for mortality forecasting. Our model shows better forecasting accuracy than the Lee–Carter and Bayesian vector autoregressive (BVAR) models without regime-switching and while retaining the advantages of BVAR. MSBVAR provides more reliable estimates for parameter uncertainty and more flexibility in the shapes of point-forecast curves and shapes of confidence intervals than BVAR. Through regime-switching, MSBVAR helps to capture transitory changes in mortality and provides insightful quantitative information about mortality dynamics. [ABSTRACT FROM AUTHOR]

This paper examines the macroeconomic effects of oil demand and supply shocks on an emerging market economy using a Bayesian vector autoregressive model combining zero and sign restrictions. The empirical analysis relies on a rich set of macroeconomic indicators mirroring the economic performance in Brazil. The effects on the Brazilian output of oil price changes driven by global demand shocks are much larger and more persistent than those driven by oil production. Oil price changes that are not related to changes in oil supply or global economic activity have a small impact on the Brazilian variables. The main policy implication of this study is that emerging market economies should identify correctly the source of oil price fluctuation for implementing the best economic policy. [ABSTRACT FROM AUTHOR]

The Greening in agro—food sector has become within the last decade a high priority issue given the 17 Sustainable targets set by OECD. More specifically, the Sustainable Development Goals (SDGs) by 2030, intend to promote using environmental resources in close correlation with measures to reduce non-environmental human pressure on the planet as well as in agro-food sector. The present work studies the greening of agro-food sector as synopsized in emissions per capita by agro-food sector for the EU and its relation to economic growth per capita with the assistance of a BVAR framework. Our findings do not validate success in greening of agro—food sector since the emissions reduction is not accompanied by economic growth a result that rejects the hypothesis of eco efficiency. Future research could involve the construction of an index that should incorporate more variables that will reflect more accurately the greening efforts in agro—food sector.

This study analyses the impact of fiscal policy on the South African economy during the period 1972Q1-2020Q2. The study adopted quarterly time series data to estimate a Bayesian Vector Autoregression (BVAR) model with the selection of hierarchical priors. The variables employed for empirical investigation included GDP, government expenditure, public debt, and gross fixed-capital formation. The results of the study show that an unexpected shock in government expenditure and public debt has a significant negative and persistent impact on economic growth in South Africa, while an unexpected shock in investment has a significant positive effect on economic growth. The findings suggest that escalating public expenditure and public debt lead to economic contraction. This implies that policy-makers ought to be cautious of excessive government expenditure and public debt to achieve fiscal consolidation. Policy-makers ought to focus on addressing structural challenges through the implementation of sound structural reform policies that aim to attract investment consistent with job creation, development and growth in South Africa’s economy.

We apply a Markov-switching Bayesian vector autoregression (MSBVAR) model to mortality forecasting. MSBVAR has not previously been applied in this context, and our results show that it is a promising tool for mortality forecasting. Our model shows better forecasting accuracy than the Lee–Carter and Bayesian vector autoregressive (BVAR) models without regime-switching and while retaining the advantages of BVAR. MSBVAR provides more reliable estimates for parameter uncertainty and more flexibility in the shapes of point-forecast curves and shapes of confidence intervals than BVAR. Through regime-switching, MSBVAR helps to capture transitory changes in mortality and provides insightful quantitative information about mortality dynamics.

This paper examines the effects of oil demand and supply shocks on emerging market economies in Latin America using a Bayesian vector autoregressive (VAR) model that combines zero and sign restrictions. Our results highlight the importance of separately identify the oil market shocks. The higher price of oil driven by increased global demand produces higher output growth in Brazil, Colombia, and Chile. The results are more persistent for Brazil and Colombia likely due to increased income from oil exports, as both economies are net oil exporters. The better times in the domestic economies result in lower uncertainty and appreciation of the exchange rate in all countries in the sample. Oil supply shocks and oil-specific demand shocks are not statistically significant for most variables. Our results provide important insights into the appropriate exchange rate policy in emerging market economies. [ABSTRACT FROM AUTHOR]

In this paper, I examine the forecasting performance of a Bayesian Vector Autoregression (BVAR) model with a steady-state prior and compare the accuracy of the forecasts against the QPM and official NBU forecasts during the Q1 2016–Q1 2020 period. My findings suggest that inflation forecasts produced by the BVAR model are more accurate than those of the QPM model for two quarters ahead and are competitive for a longer time horizon. The BVAR forecasts for GDP growth also outperform those of the QPM but for the whole forecast horizon. Moreover, it is revealed that the BVAR model demonstrates a better performance compared to the NBU’s official inflation forecasts over the monetary policy horizon, whereas the opposite is true for GDP growth forecasts. Future research may deal with estimation issues brought about by COVID-19.

This study analyses the impact of fiscal policy on the South African economy during the period 1972Q1-2020Q2. The study adopted quarterly time series data to estimate a Bayesian Vector Autoregression (BVAR) model with the selection of hierarchical priors. The variables employed for empirical investigation included GDP, government expenditure, public debt, and gross fixed-capital formation. The results of the study show that an unexpected shock in government expenditure and public debt has a significant negative and persistent impact on economic growth in South Africa, while an unexpected shock in investment has a significant positive effect on economic growth. The findings suggest that escalating public expenditure and public debt lead to economic contraction. This implies that policy-makers ought to be cautious of excessive government expenditure and public debt to achieve fiscal consolidation. Policy-makers ought to focus on addressing structural challenges through the implementation of sound structural reform policies that aim to attract investment consistent with job creation, development and growth in South Africa's economy. [ABSTRACT FROM AUTHOR]

Major equipping and allocation of credits in the Iranian economy is carried out by the banking system. Therefore, considering the role of banks in the economy, especially the financing of economic enterprises, the impact of banking facilities through various contracts on the phenomenon of stagflation is essential. The bank credit channel is one of the channels of monetary transfer in the mechanism of monetary policy transfer, and monetary policy of the central bank is implemented through a variety of contracts through a bank crediting channel. In this research, the effect of Participative contracts, exchangeable contracts, direct investment and Qarzol-Hasaneh contracts in free usury (interest) banking on Stagflation in Iran has been studied. For this purpose, by Using the seasonal data of Islamic-based facilities along with exchange rate, oil revenues and government expenditures variables during the period of 2000-2016, within the framework of the BVAR model and the Normal-Wishart prior of Sims-Zha has been evaluated. The results show that increasing the level of facilities within the framework of Participative contracts, direct investment and Qarzol-Hasaneh contracts leads to a relative and limited increase of stagflation, and in contrast to the exchangeable contracts, leads to a reduction of stagflation.

Given the importance of financing in investment projects and different ways of financing, the choice between financing methods can be focused on the type of impact of each on growth production. For this purpose, in this study, sources such as foreign investment, foreign debt, oil and non-oil revenues as financing methods in Iran have been studied. In the first step, the BVAR model of analysis based on analysis of variance and instantaneous reaction functions was used to analyse the quantity and quality of the response of the model variables to each other and also the reaction of the main variable (economic growth) to other variables. In the second step, the ARDL model is used to analyse the model dynamics. The research period is 1979 to 2016 and the data are annual. The results show that overall financing by oil sales, foreign investment and foreign debt have a positive effect on economic growth, emphasizing that the effect of oil revenues is greater and faster than others.

This paper explores the transmission mechanism between wages and prices in the euro area. While labour costs account for notable parts of companies' costs and should, hence, be of relevance for the developments in inflation, the empirical literature suggests only a loose link between the variables. From a theoretical perspective there are at least two broad categories of channels of how labour costs and prices can be linked, namely the supply-side channel of cost push inflation and the demand-side channel of demand pull inflation. This paper develops a BVAR model for the wage-price pass-through in the euro area and analyses how supply and demand shocks that trigger these two types of inflation are transmitted between wages and prices along the cost side of the economy. The main results can be summarized as follows: First, the wage-price pass-through in the euro area is shock-dependent. Second, the pass-through of wages to prices appears to have undergone some changes over time. Third, the pass-through appears to differ across HICP components. Finally, historical decompositions show that the identified shocks are of relevance for the movements of wages and prices in the euro area. [ABSTRACT FROM AUTHOR]

Inflation expectations are an important marker for the conduct of monetary policy. Using a Bayesian structural VAR‐X model we analyze the macroeconomic factors that determine inflation expectations of the general public in India. Besides the standard macroeconomic factors such as output, inflation rate, monetary policy, and exchange rate, we include economic policy uncertainty as an endogenous variable, while international factors such as oil prices and financial volatility that have spill‐over effects are included as exogenous variables. We find that oil prices and the exchange rate have considerable effects on expectations of inflation, and that in a longer horizon, international financial volatility also matters. Our model generates results free from the "price puzzle" and the "exchange rate puzzle" that otherwise exist in the literature on monetary transmission mechanisms for India. [ABSTRACT FROM AUTHOR]

Can information on macroeconomic uncertainty improve the forecast accuracy for key macroeconomic time series for the US? Since previous studies have demonstrated that the link between the real economy and uncertainty is subject to nonlinearities, I assess the predictive power of macroeconomic uncertainty in both linear and nonlinear Bayesian VARs. For the latter, I use a threshold VAR that allows for regime-dependent dynamics conditional on the level of the uncertainty measure. I find that the predictive power of macroeconomic uncertainty in the linear VAR is negligible. In contrast, using information on macroeconomic uncertainty in a threshold VAR can significantly improve the accuracy of short-term point and density forecasts, especially in the presence of high uncertainty. [ABSTRACT FROM AUTHOR]

Este artículo analiza el efecto de distintos choques monetarios y externos sobre el crecimiento cíclico del departamento del Valle del Cauca y se compara con el efecto que tienen estos sobre el ciclo nacional en el periodo comprendido entre 2013 y 2016, en el cual caen los precios de las materias primas. La estrategia empírica empleada permite examinar la efectividad de la política monetaria y cambiaria en la región. Las estimaciones muestran que el mejor desempeño económico del Valle, en comparación con el promedio nacional, se debe, principalmente, a la mayor sensibilidad a la tasa de cambio real y a la menor dependencia del precio del petróleo.

• An expansionary monetary policy shock does not have a significant effect on income inequality. • Expansionary monetary policy increase wealth inequality mainly through the portfolio channel. • There are several transmission channels that have an impact in the opposite direction that leads to an uncertain net effect. • The redistributive effects of monetary policy should be considered in order to design properly such policies. We assess the effects of monetary policy shocks on income and wealth inequality through direct inequality measures and by analyzing several transmission channels explored in recent literature. Furthermore, we analyze two additional channels: the Housing and the Fiscal channels. The methodology adopted is a Bayesian proxy SVAR using a high-frequency identification based on the external instruments approach. Our own policy shocks are constructed for this purpose. The results show that an expansionary monetary policy shock does not have a significant effect on income inequality due to the existence of opposite channels, whereas it increases wealth inequality mainly through the portfolio channel. [ABSTRACT FROM AUTHOR]

During the crisis of 2009–2013 many institutes made large errors in their unemployment forecasts. This paper develops alternative short-run models to improve these forecasts. The models are applied to the Netherlands and compared to the unemployment forecasts of the CPB. A BVAR model performs significantly better than the CPB forecasts. The BVAR can also be used to predict the labor market transition probabilities used in the forecast functions of two-state and three-state stock-flow models of unemployment. This also results in significantly better forecasts. However the combination of these three best-performing models produces the largest gain in forecast accuracy. [ABSTRACT FROM AUTHOR]

The transition to a model-based forecasting environment is encountered by hurdles in a small open developing economy. An attempt is made to validate the benefits of model-based inflation forecasting for central banks in small open developing economies. Despite data limitations, two distinct VARs are designed to project near-term inflation. Batteries of tests (such as sequential forecasts, out-of-sample forecasting errors, equal-weight forecasting errors and decomposition) are performed on the two models to assess their predictive ability. The main finding is that model-based forecasts are reliable for use by central banks in small open developing economies, as substantiated by the relatively low forecasting errors. [ABSTRACT FROM AUTHOR]

GDP forecasting remains a challenge for a small open developing economy. Faced with insufficient and low-frequency data, central bank forecasters cannot project GDP reliably for the purpose of monetary policy decision-making. An attempt is made to forecast GDP using a factor-augmented vector autoregressive (FAVAR) model for a small open developing economy. The forecasting accuracy of the FAVAR model is examined through sequential forecasts and benchmarked against a Bayesian vector autoregressive (BVAR) model. The main finding of this study is that a FAVAR model can generate consistent GDP projections for a small open developing economy despite data inadequacy. [ABSTRACT FROM AUTHOR]

In this paper the interaction between the production macroeconomic indicators of the Russian economy and MIBOR (the main operational benchmark of the Bank of Russia), as well as the relationship between the inflation indicators and money supply were investigated with Bayesian approach. Conjugate Normal Inverse Wishart Prior was used. According to the study, tough monetary policy has a deterrent effect on the Russian economy. The growth of the money market rate causes a reduction in investments and output in the main sectors of the economy, as well as a drop in the income of the population with an increase in the unemployment rate.

This paper provided a full characterization of several monetary aggregates over Portuguese's historical economic business cycles. By focusing on the 1911–1999 period, the paper also revisits the issue of the role of money on real macroeconomic outcomes, inspired from the monetarists versus Keynesians debate on quest for validity of money (non-)neutrality. By means of descriptive statistics we first uncover that money changes were associated with changes in real economic activity. Most monetary aggregates are more volatile than GDP, display high serial autocorrelation, are generally countercyclical and lead the economic cycle. Then, through econometric analysis, our results show that our monetary series were characterized by unit roots and were cointegrated with real GDP. Evidence also suggested that money supply Granger-caused real GDP. Finally, both variance decomposition and impulse response function analyses from an estimated BVAR, uncovered a persistent and mutual effect running from a shock in real GDP to monetary aggregates and vice versa, therefore supporting the money non-neutrality hypothesis in the case of Portugal. [ABSTRACT FROM AUTHOR]

This paper examines whether the transmission mechanism of monetary policy in South Africa has changed after the global financial crisis (GFC). We use a Bayesian vector autoregressive (BVAR) model with Minnesota priors and 15 monthly variables, extending the system of Christiano, Eichenbaum, with Evans (1999). The benefit of the BVAR approach is that it can accommodate a large cross section of variables without running out of degrees of freedom. To identify the change in the transmission process, we divide the sample size into two subsamples, namely the pre‐GFC period (March 2001 to August 2008) and the post‐GFC period (September 2008 to February 2016). The results indicate that a change in the transmission of monetary policy occurred after the GFC. The magnitude of the effect of a monetary policy shock on output is considerably greater in the pre‐GFC period compared to the post‐GFC period. Moreover, the impact of a policy shock on inflation is not statistically significant in the post‐GFC period. The variance decomposition shows that the interest‐rate channel has possibly weakened in the post‐GFC period. [ABSTRACT FROM AUTHOR]

The monetary analysis based on the BVAR (Bayesian Vector Autoregression) model is extremely important in the monetary policy implementation strategy, the information provided is important not only for the Central Bank, but also for the economic agents and the population. Therefore, conducting this analysis at the level of Romania helps to understand better the mechanism by which monetary policy is transmitted in order to achieve the set target, namely inflation targeting, but it also provides us with important information regarding the accession to the euro area. The model we are trying to test helps us understand through the correlations between the interest rate, GDP and the inflation rate how monetary policy responds to shocks. The model follows the methodology presented by Sims and Zha (1998) in the paper "Bayesian Methods for Dynamic Multivariate Models and Using the Bayesian Autoregressive Vector". In the analysis of this model, quarterly data for a minimum of three years, three variables are used to make the results relevant. The data needed to model the model are used in logarithmic form, except for the interest rate, and the outcome is applied to a differentiated premium operator. Of the variables used, the interest rate is the only one that does not allow seasonal adjustment.

One of the promising approaches of macroeconomic modeling and quantitative assessment of the impact of external and internal factors on macroeconomy of a country, which is actively used abroad, is a Bayesian approach to the estimation of vector autoregressions. In this paper we examine Bayesian vector autoregression model (BVAR) to assess the impact of external shocks, such as the price of Brent crude oil, the volatility index VIX and the Shanghai Stock Exchange Composite index, on Russian macroeconomic indicators. The results allow us to estimate the contribution of external factors as a significant in the dynamics of Russia economic variables. This approach can be successfully applied for the analysis of Russian data, which was confirmed by the results presented in the article.

Este artículo analiza los posibles efectos de los diferentes choques de los sectores financiero, fiscal y externo sobre la evolución de la actividad macroeconómica de Colombia. A partir de un enfoque analítico de demanda agregada, se emplean modelos VAR Bayesianos utilizando la densidad a priori de Minnesota, con datos trimestrales para el periodo 1995-2013. A través de los análisis de impulso-respuesta y causalidades de Granger, los resultados muestran los posibles escenarios de riesgo en el crecimiento económico agregado, así como hechos estilizados respecto a relaciones entre variables de interés.

Reflecting the importance of commodities for the Australian economy, we construct a dynamic stochastic general equilibrium (DSGE) model of the Australian economy with a commodity sector. We assess whether its forecasts can be improved by using it as a prior for an empirical Bayesian vector autoregression (BVAR). We find that the forecasts from the BVAR tend to be more accurate than those from the DSGE model. Nevertheless, for output growth these forecasts do not outperform benchmark models, such as a small open economy BVAR estimated using the standard priors for forecasting. A Bayesian factor augmented vector autoregression produces the most accurate near-term inflation forecasts. [ABSTRACT FROM PUBLISHER]

ABSTRACT This paper proposes a large Bayesian Vector Autoregressive (BVAR) model with common stochastic volatility to forecast global equity indices. Using a monthly dataset on global stock indices, the BVAR model controls for co-movement commonly observed in global stock markets. Moreover, the time-varying specification of the covariance structure accounts for sudden shifts in the level of volatility. In an out-of-sample forecasting application we show that the BVAR model with stochastic volatility significantly outperforms the random walk both in terms of point as well as density predictions. The BVAR model without stochastic volatility, on the other hand, shows some merits relative to the random walk for forecast horizons greater than six months ahead. In a portfolio allocation exercise we moreover provide evidence that it is possible to use the forecasts obtained from our model with common stochastic volatility to set up simple investment strategies. Our results indicate that these simple investment schemes outperform a naive buy-and-hold strategy. [ABSTRACT FROM AUTHOR]

This paper discusses how the forecast accuracy of a Bayesian vector autoregression (BVAR) is affected by introducing the zero lower bound on the federal funds rate. As a benchmark I adopt a common BVAR specification, including 18 variables, estimated shrinkage, and no nonlinearity. Then I entertain alternative specifications of the zero lower bound. I account for the possibility that the effect of monetary policy on the economy is different in this regime, replace the federal funds rate by its shadow rate, consider a logarithmic transformation, feed in monetary policy shocks, or utilize conditional forecasts allowing for all shocks implemented through a rejection sampler. The latter two are also coupled with interest rate expectations from future contracts. It is shown that the predictive densities of all these specifications are greatly different, suggesting that this modeling choice is not innocuous. The comparison is based on the accuracy of point and density forecasts of major US macroeconomic series during the period 2009 : 1 to 2014 : 4. The introduction of the zero lower bound is not beneficial per se, but it depends on how it is done and which series is forecasted. With caution, I recommend the shadow rate specification and the rejection sampler combined with interest rate expectations to deal with the nonlinearity in the policy rate. Since the policy rate will remain low for some time, these findings could prove useful for practical forecasters. [ABSTRACT FROM AUTHOR]

The monetary analysis based on the BVAR (Bayesian Vector Autoregression) model is extremely important in the monetary policy implementation strategy, the information provided is important not only for the Central Bank, but also for the economic agents and the population. Therefore, conducting this analysis at the level of Romania helps to understand better the mechanism by which monetary policy is transmitted in order to achieve the set target, namely inflation targeting, but it also provides us with important information regarding the accession to the euro area. The model we are trying to test helps us understand through the correlations between the interest rate, GDP and the inflation rate how monetary policy responds to shocks. The model follows the methodology presented by Sims and Zha (1998) in the paper "Bayesian Methods for Dynamic Multivariate Models and Using the Bayesian Autoregressive Vector". In the analysis of this model, quarterly data for a minimum of three years, three variables are used to make the results relevant. The data needed to model the model are used in logarithmic form, except for the interest rate, and the outcome is applied to a differentiated premium operator. Of the variables used, the interest rate is the only one that does not allow seasonal adjustment. [ABSTRACT FROM AUTHOR]

We analyse the effect of the European Central Bank's (ECB) monetary shock spillover and its impact on the European Union's 9 countries outside the euro area (EA) between 2000 and 2020. We use the sign-restricted Bayesian VAR model and subsequent interpretation by plotting the impulse-response functions. Moreover, we investigate both conventional and unconventional monetary policies and its international transmission. The spillover of monetary policy is growing with the openness of economies and the ongoing deepening of integration. The output responds to the EA's monetary shock flexibly and persistently, but there is considerable heterogeneity across countries. We claim that it is essential for central banks outside the EA to monitor and incorporate the ECB's monetary policy spillover into decision-making processes. In particular, the international transmission of the unconventional monetary policy has a fundamental effect on the development of the price level, thus achieving price stability. In the case of implementing a counter-cyclical policy, it is also necessary to monitor conventional policy. However, there is no need to fight the spillover effect since there is no beggar-thy-neighbour problem, i.e., spillover effect works in the same direction in both domestic and foreign country. [ABSTRACT FROM AUTHOR]

Mexico has become one of the top tourist destinations in the world, which is reflected in the high places obtained in the international rankings. However, the Mexican tourism sector has presented a strong dependence on the US economy. That is why in this work by using a Structural Vector Autoregressive Model, SVAR, we examine and measure the impact of the US economic growth and the exchange rate Peso-American Dollar on the total visitor arrivals to Mexico. We have found that both variables have had a positive and a statistically significant influence on the Mexican tourism sector. [ABSTRACT FROM AUTHOR]

It will be beneficial to devise an effective approach for short-term macroeconomic forecasting. Existing traditional statistics-based macroeconomic forecasting mainly focuses on exploring feasible methods for improving the accuracy of long-term predictions. However, the performance of short-term predictions was far less impressive under the impact of unexpected incidents. Furthermore, some deep learning-based approaches can achieve fine-grained variable frequency forecasting and preliminarily demonstrate effective results, but the interpretability is still controversial. Therefore, how to consider both the performance and the interpretability has already become a universal concern and an urgent unsolved problem. In this paper, we identified the above issue and proposed an interpretable data-driven approach, named EcoForecast 1 1 The code and data are available at https://github.com/navfour/ecoforecast. , for short-term macroeconomic forecasting based on the N-BEATS (neural basis expansion analysis for interpretable time series forecasting) neural network. To the best of our knowledge, EcoForecast is the first interpretable purely data-driven unified normative scheme for macroeconomic forecasting that achieves variable forecast frequencies and prediction domains, surpassing traditional statistics-based and deep learning-based approaches in performance or interpretability. EcoForecast used a three-level hierarchical signal encoding, including the fully connected neural network (FCNN) level, the block level, and the stack level. The FCNN level implemented both forecast and backcast information extraction for the temporal prediction and parameter learning of context. Block levels were connected by residuals so that the block's backcast could be sequentially filtered on the input. The stack level was used to form the top-level system of EcoForecast, where each stack was constrained to specialize in different inductive functions. To some extent, EcoForecast can balance effectiveness, efficiency, generalizability and interpretability while conditions such as forecast frequency and time window change, even when unexpected incidents occur. Based on the actual macroeconomic data for China from 1992 to 2022, the data-driven EcoForecast demonstrated high stability in different sequence learning scenarios and the accompanying high-accuracy performance. This stability was reflected in smaller prediction error expectation and variance, tolerance of fewer input samples, and robustness across prediction domains. The experimental results indicated that EcoForecast improved the accuracy up to 3.94 times compared with the traditional BVAR. In the robustness test, EcoForecast required only a quarter of the data to achieve 2.51 times smaller forecast errors than the BVAR while also improving the accuracy of varied macroeconomic indicators such as the Purchasing Managers' Index (PMI) and national electricity generation (ELEC)forecasting by 2.38 and 1.45 times. EcoForecast had a high sensitivity to the emergence of economic inflection points and adapts quickly when the economic environment changes, thus demonstrating a performance that exceeds traditional solutions in GDP forecasting during epidemics and PMI forecasting during economic turmoil. Interfacing with traditional economics research, the interpretable EcoForecast can uncover the trends and cycles of economic change, from which the conclusions are validated with the actual economics practice in China. Our findings can provide a new possible research direction for short-term macroeconomic forecasting. [ABSTRACT FROM AUTHOR]