Macroeconomic Time Series Data Research Articles

Estimating Posterior Sensitivities with Application to Structural Analysis of Bayesian Vector Autoregressions

The inherent feature of Bayesian empirical analysis is the dependence of posterior inference on prior parameters, which researchers typically specify. However, quantifying the magnitude of this dependence remains difficult. This article extends Infinitesimal Perturbation Analysis, widely used in classical simulation, to compute asymptotically unbiased and consistent sensitivities of posterior statistics with respect to prior parameters from Markov chain Monte Carlo inference via Gibbs sampling. The method demonstrates the possibility of efficiently computing the complete set of prior sensitivities for a wide range of posterior statistics, alongside the estimation algorithm using Automatic Differentiation. The method’s application is exemplified in Bayesian Vector Autoregression analysis of fiscal policy in U.S. macroeconomic time series data. The analysis assesses the sensitivities of posterior estimates, including the Impulse response functions and Forecast error variance decompositions, to prior parameters under common Minnesota shrinkage priors. The findings illuminate the significant and intricate influence of prior specification on the posterior distribution. This effect is particularly notable in crucial posterior statistics, such as the substantial absolute eigenvalue of the companion matrix, ultimately shaping the structural analysis.

Revisiting the relationship between oil supply news shocks and U.S. economic activity: Role of the zero lower bound

Do oil supply news shocks affect the U.S. economy differently at the zero lower bound? We provide evidence consistent with this hypothesis. Using various U.S. macroeconomic time series data from 1975, we find that the transmission of oil price shocks driven by changes in supply expectations is substantially different at the zero lower bound. These shocks tend to be contractionary when interest rates are higher, and expansionary when monetary policy is operating in the ZLB. Private business investment plays a key role in altering the transmission of oil supply news shocks around the ZLB. The results continue to hold even after controlling for a potentially confounding non-linearity associated with the U.S. shale oil boom. Our empirical investigation should improve our understanding of the role played by policy and structural factors in altering the transmission of oil shocks to the U.S. economy.

Macroeconomic and financial mixed frequency factors in a big data environment

Abstract In this paper, we evaluate the predictive content of 3 new business condition indexes and uncertainty measures that are estimated using high-frequency financial and low-frequency macroeconomic time series data. More specifically, our measures are defined as latent factors that are extracted from a state space model that includes multiple different frequencies of non-parametrically estimated components of quadratic variation, as well as mixed frequency macroeconomic variables. When forecasting growth rates of various monthly financial and macroeconomic variables, use of our new mixed frequency factors is shown to result in significant improvement in predictive performance, relative to a number of benchmark models. Additionally, when used to forecast corporate yields, predictive gains associated with the use of our measures are shown to be monotonically increasing, as one moves from predicting higher to lower rated bonds. This is consistent with the existence of a natural pricing channel wherein financial risk (as measured using our volatility factors) contains more predictive information for lower grade bonds. We also find that a variety of extant risk factors including the Aruoba et al. [(2009a). Real-time measurement of business conditions. Journal of Business & Economic Statistics, 27(4), 417427] business conditions index also contain marginal predictive content for the variables that we examine, although their inclusion does not reduce the usefulness of our measures.

Renewable Energy Mix Enhancement: The Power of Foreign Investment and Green Policies

The objective of this study is to examine and assess the viewpoint of population and economic expansion, as well as foreign investment in the renewable energy sector, by examining the influence of green investment on indexed listed businesses in Indonesia's Sri-Kehati. The dataset utilised in this study include macroeconomic time series data from Indonesia spanning the years 2015 to 2021, as well as issuers indexed under the SRI-KEHATI framework. The data underwent analysis using the multiple regression approach. The study employed an associative research technique and utilised the Eviews 10 programme tool for data analysis. The findings of the study demonstrate a noteworthy positive impact of economic growth on the composition of renewable energy sources. The impact of population growth on the composition of renewable energy sources is shown to be insignificant, but foreign investment is observed to have a notable and adverse influence on the renewable energy mix. The impact of foreign investment on the Renewable Energy Mix is found to be both statistically significant and unfavourable. The variable of economic growth exhibits a noteworthy and favourable impact when moderated by green investment in Sri-Kehati indexed issuers, specifically in relation to the composition of renewable energy sources. The variable of population growth exhibits a noteworthy and favourable impact by means of moderating green investment in the perspective of Sri-Kehati indexed issuer companies regarding the composition of renewable energy sources. The variable of economic growth demonstrates a statistically significant and favourable impact when moderated by green investment in the Sri-Kehati Indexed Issuer Company, specifically in relation to the renewable energy mix. The impact of population growth on the renewable energy mix in the perspective of Sri-Kehati indexed issuer companies is found to be positive and statistically significant when considering the moderation effect of green investment. The variable representing economic growth demonstrates a notable and favourable impact when moderated by green investment in the Sri-Kehati Indexed Issuer Company, particularly in relation to the composition of renewable energy sources. The impact of population growth on the renewable energy mix in the perspective of Sri-Kehati indexed issuer companies is found to be positive and statistically significant when considering the moderation effect of green investment.

Government expenditure structure, technological progress and economic growth

PurposeThis study examines the differences in the economic stimulus effects, transmission mechanisms, and output multipliers of government consumption, government traditional investment, and government science and technology investment.Design/methodology/approachThis study constructs and estimates a New Keynesian model of endogenous technological progress embedded in the research and development (R&D) and technology transfer sectors. Using Chinese macroeconomic time series data from 1996 to 2019, this study calibrates and estimates the model and analyzes the impulse response function and a counterfactual simulation of expenditure structure adjustment.FindingsThe results show that compared with the traditional dynamic stochastic general equilibrium (DSGE) model, the endogenous process of technological progress amplifies the impact of government consumption shock and traditional government investment shock on the macroeconomy, leading to greater economic cycle fluctuations. As government investment in science and technology has positive external spillover effects on firm R&D activities and the application of innovation achievements, it can promote more sustainable economic growth than government consumption and traditional investment in the long run.Originality/valueThis study constructs an extended New Keynesian model with different types of government spending, which includes endogenous technological progress within the R&D and technology transfer sectors, thereby linking fiscal policy, business cycle fluctuations and long-term economic growth. This model can study the macroeconomic impact of fiscal expenditure structure adjustment when fiscal expansion is limited. In the Bayesian estimation of model parameters, this study not only uses macroeconomic variables but also adds a sequence of private R&D investment.

Bayesian optimization of hyperparameters from noisy marginal likelihood estimates

SummaryBayesian models often involve a small set of hyperparameters determined by maximizing the marginal likelihood. Bayesian optimization is an iterative method where a Gaussian process posterior of the underlying function is sequentially updated by new function evaluations. We propose a novel Bayesian optimization framework for situations where the user controls the computational effort and therefore the precision of the function evaluations. This is a common situation in econometrics where the marginal likelihood is often computed by Markov chain Monte Carlo or importance sampling methods. The new acquisition strategy gives the optimizer the option to explore the function with cheap noisy evaluations and therefore find the optimum faster. The method is applied to estimating the prior hyperparameters in two popular models on US macroeconomic time series data: the steady‐state Bayesian vector autoregressive (BVAR) and the time‐varying parameter BVAR with stochastic volatility.

Including news data in forecasting macro economic performance of China

In this work we predict changes in the Gross Domestic Product (GDP) of China using dynamic factor models. We report results of 3- and 6-months ahead forecasts, where we use $$124$$ predictors from various sources and dates ranging from 2000 through 2017. Our analysis includes China specific macroeconomic time series data and a large number of predictor variables. We follow the latest state of the art, as outlined by, Stock and Watson (in: Handbook of macroeconomics vol 2, Elsevier, pp 415–525, 2016) who use principal component analysis (PCA) to reduce number of variables and apply dynamic factor model (DFM) to make predictions. The results suggest that including news sentiment significantly improves forecasts and this approach outperforms univariate autoregression. The contributions of this paper are two fold, namely, the use of news to improve forecasts and superior forecast of China’s GDP.

Weighted least squares estimation in a binary random coefficient panel model with infinite variance

This article investigates the asymptotic properties of weighted least squares estimators (WLSE) for a binary random coefficient autoregressive (RCA) panel model with heterogeneous variances of panel variables. It is an extension of Johansen and Lange (2013) to a panel model, which is more practical for macroeconomic time series data. We develop asymptotic properties of the WLSE in cases of finite and infinite variances, respectively, as both sizes of panels and samples tend to infinity. In the latter case with infinite variance, the asymptotic for the WLSE βˆ of the coefficient β is shown to be a curious result βˆ⟶pβ−1. It is proven by using the notion of a tail index and the stable distribution limit. In a Monte Carlo simulation, feasible WLSEs are computed iteratively and some evidences are given to verify our theoretical results.

Wisdom of Malaysia’s Guided Economic Globalization in the Milieu of Today’s Capitalist Globalization

Capitalist globalization is a project of market dogmatism founded on the Darwinist social philosophy of ‘survival of the fittest’. In the political economy of the world with unequal nations and imperfect competition, it is a suspect of deceptive game for the developing countries. This paper is an attempt to analyze the wisdom of Malaysia’s guided economic globalization strategy using the technique of log-linear pooled ordinary least square regression and related macroeconomic time series data for the period 1982-2016. The empirical results suggest that Malaysia’s guided globalization strategy has been found successful in achieving most of its national macroeconomic goals. But in terms of international economic competitiveness the success has been only modest, its per capita income ratios with its two most important trading and development partners - USA and Japan - are found decreasing steadily having a catching up phenomenon along the path of income convergence. However, its per capita income gaps with them in absolute term are found still large and widening to suggest that developing countries must pursue globalization very cautiously and selectively.

Including News Data in Forecasting Macro Economic Performance of China

In this paper we report results of 3- and 6-months ahead forecasts of Gross Domestic Product (GDP) of China. In total we use 124 predictors from various sources and dates ranging from 2000 through 2017. We use China specific macroeconomic time series data and a large number of predictor variables. In our study we follow the latest state of the art, as outlined by, [Stock and Watson, 2016] who use principal component analysis (PCA) to reduce number of variables and apply dynamic factor model (DFM) to make predictions. The results suggest that including news sentiment significantly improves forecasts and this approach outperforms univariate autoregression. The contributions of this paper are two fold, namely, the use of news to improve forecasts and superior forecast of China's GDP.

Automated Sensitivity Computations for Bayesian Markov Chain Monte Carlo Inference: A New Approach for Prior Robustness and Convergence Analysis

Infinitesimal perturbation analysis is a widely used approach to assess the input sensitivities of stochas- tic dynamic systems in the classical simulation context. In this paper, we introduce an efficient nu- merical approach to undertake Infinitesimal perturbation analysis in the context of dependent Markov Chain Monte Carlo simulations widely used in Bayesian inference. Building on recent developments, we develop a scheme based on automatic differentiation to compute posterior sensitivities based on exact (up to computer floating point error) first-order derivatives of draws (Jacobians) with respect to prior input parameters alongside the estimation algorithm. Assessing such local robustness of posterior inference poses a challenge for existing methods such as finite differencing, symbolic differentiation, and likelihood ratio methods due to the complex stochastic dependence structure and computational intensity of dependent sample draws. The proposed methods allow for a comprehensive and exact local sensitivity analysis of Markov Chain Monte Carlo output with respect to all input parameters, i.e., prior hyper-parameters (prior robustness) and chain starting values (convergence). Our imple- mentation focuses on methods for Gibbs-based algorithms widely used for inference and forecasting in complex high-dimensional settings. We illustrate how the methods can help practitioners assess convergence and prior robustness in an application of Bayesian Vector Autoregression with shrinkage priors for US macroeconomic time series data and forecasting.

Small and medium enterprises led-growth in two Adriatic countries: Granger causality approach

This paper investigates causal relationships between gross domestic product (GDP) and the number of small and medium sized enterprises (SMEs) controlled for unemployment rates (UNR) within Slovenia and Croatia. Macro-economic time series data on GDP, SMEs and UNR are analysed in a unit root framework and applied regression analysis. These processes are known as the Johansen co-integration test and Granger-Causality-test. The results show that UNR, GDP and SMEs did not have causal relationships in Croatia between January 2008 and December 2013. UNR and SMEs have a bidirectional relationship in Slovenia with a greater number of SMEs per capita than in Croatia. During their economic recovery period between January 2014 and December 2017, Slovenia and Croatia have experienced the causal unidirectional relation from SMEs to GDP as a positive signal under seen to policy makers on usefulness of investment in SMEs during economic prosperity. The results of a vector autoregressive model suggest a 1% change in the number of SMEs in Croatia decreases GDP in the time of crisis by almost 1.8%. For Slovenia, there is no statistically significant cointegration vector pertained to SMEs-led growth. Finally, the unidirectional causality relation from SMEs to UNR is statistically significant for Slovenia.

UNDERSTANDING INDONESIA’S MACROECONOMIC DATA: WHAT DO WE KNOW AND WHAT ARE THE IMPLICATIONS?

Unit root properties of macroeconomic data are important for both econometric modelling specifications and policy making. The form of variables (whether they are a unit root process) helps determine the correct econometric modelling. Equally, the form of variables helps explain how they react to shocks (both internal and external). Macroeconomic time-series data are often at the forefront of shock analysis and econometric modelling. There is a growing emphasis on research on Indonesia using time-series data; yet, there is limited understanding of data characteristics and shock response of these data. Using an extensive dataset comprising 33 macroeconomic time-series variables, we provide an informative empirical analysis of unit root properties of data. We find that regardless of data frequencies the empirical evidence of unit roots is mixed, some series respond quickly to shocks others do take time, and almost every macroeconomic data suffers from structural breaks. We draw implications of these findings.

Has Financial Liberalisation Promoted Economic Growth in Nigeria? Evidence from Auto-Regressive Distributed Lag (ARDL) Approach

This study attempted to ascertain whether or not financial liberalisation policies promoted economic growth in Nigeria for the period spanning 1986 –2014, using Auto-regressive Distributed Lag (ARDL)-bounds testing approach and unrestricted error correction model (UECM) to cointegration analysis. Macroeconomic time series data were obtained from Central Bank of Nigeria (CBN) Annual Statistical Bulletins, National Bureau of Statistics (NBS) Bulletins, and World Bank Development Indices. Three alternative measures of financial liberalisation (FINDEX1, FINDEX2, FINDEX3) and six intervening variables were incorporated into the study. The empirical findings show that the impact of financial liberalization on economic growth varied for different measures of financial liberalization undertaken; and were significant both in the short run and the long run: FINDEX1 promoted economic growth significantly, and corroborates the growth-stimulating effect of financial liberalization; while FINDEX2 and FINDEX3 exerted significant negative impacts on economic growth. Real exchange rate, degree of openness, consumer price index, increase in life expectancy of labour, and debt service ratio significantly influenced economic growth in the short run and long run. None of the measures of financial liberalisation Granger-caused economic growth, but a one-way causation ran from FINDEX2 to FINDEX1, and from FINDEX3 to FINDEX1.The study recommends increased financial deepening, reduction of lending rates, removal of bottlenecks in the banking and stock markets institutions in Nigeria.

Deciding between alternative approaches in macroeconomics

Macroeconomic time-series data are aggregated, inaccurate, non-stationary, collinear and rarely match theoretical concepts. Macroeconomic theories are incomplete, incorrect and changeable: location shifts invalidate the law of iterated expectations and ‘rational expectations’ are then systematically biased. Empirical macro-econometric models are non-constant and mis-specified in numerous ways, so economic policy often has unexpected effects, and macroeconomic forecasts go awry. In place of using just one of the four main methods of deciding between alternative models, theory, empirical evidence, policy relevance and forecasting, we propose nesting ‘theory-driven’ and ‘data-driven’ approaches, where theory-models’ parameter estimates are unaffected by selection despite searching over rival candidate variables, longer lags, functional forms, and breaks. Thus, theory is retained, but not imposed, so can be simultaneously evaluated against a wide range of alternatives, and a better model discovered when the theory is incomplete.

Multiple output regression with latent noise

In high-dimensional data, structured noise caused by observed and unobserved factors affecting multiple target variables simultaneously, imposes a serious challenge for modeling, by masking the often weak signal. Therefore, (1) explaining away the structured noise in multiple-output regression is of paramount importance. Additionally, (2) assumptions about the correlation structure of the regression weights are needed. We note that both can be formulated in a natural way in a latent variable model, in which both the interesting signal and the noise are mediated through the same latent factors. Under this assumption, the signal model then borrows strength from the noise model by encouraging similar effects on correlated targets. We introduce a hyperparameter for the latent signal-to-noise ratio which turns out to be important for modelling weak signals, and an ordered infinite-dimensional shrinkage prior that resolves the rotational unidentifiability in reduced-rank regression models. Simulations and prediction experiments with metabolite, gene expression, FMRI measurement, and macroeconomic time series data show that our model equals or exceeds the state-of-the-art performance and, in particular, outperforms the standard approach of assuming independent noise and signal models.

Growth in a time of austerity: evidence from the UK

AbstractThis paper uses an empirical approach to test the specific causal relationship between debt and growth in the UK, in the context of the debate surrounding the use of a policy known as austerity measures. This time series perspective makes use of more recent Granger causality and cointegration tests that allow for non‐stationarity in macroeconomic time series data in the presence of structural breaks. Controlling for exogenous shocks associated with the period around the financial crisis, we find no evidence of a causal relationship between economic growth and public debt for the UK.

Entrepreneurial Impulse, Investment Behavior, and Economic Fluctuations—A VAR Analysis with Indian Data

This study analyzes the observed behavior of growth cycles and the dynamics of economic fluctuations in terms of entrepreneurial impulse with Indian macroeconomic time series data for more than 60 years. The fluctuations of investment are explained in terms of persistence, volatility, and comovements of the cyclical components. The study observes that the growth cycles of private investment were much more volatile than the growth cycles of public investment and gross domestic product. Investments in India were procyclical, but their growth cycles became acyclical. This study observes that the fluctuations of growth cycle frequencies of private investment appeared even when there were no shocks to gross domestic product. Such investment behavior may be because of the self-fulfilling beliefs of investors. Private investment in India has not been badly affected by the severe balance of payments crises. Rather, a cyclical downturn was seen as an opportunity to invest by the large business houses.

Real Rigidities and the Cross-Sectional Distribution of Price Stickiness: Evidence from Micro and Macro Data Combined

We use a standard sticky-price model to provide evidence on three mechanisms that can reconcile somewhat frequent price changes with large and persistent real effects of monetary shocks. To that end, we estimate a semi-structural model for the U.S. economy that allows for varying degrees of real rigidities, and cross-sectional heterogeneity in price stickiness. The model can extract some information about these two features of the economy from aggregate data, and discriminate between different distributions of price stickiness. Hence it can also speak to the debate about the role of sales and other temporary price changes in shaping the effects of monetary policy. Employing a Bayesian approach, we combine macroeconomic time-series data with information about empirical distributions of price stickiness derived from micro price data for the U.S. economy. Our estimates point to the presence of both large real rigidities and an important degree of heterogeneity in price stickiness. Moreover, cross-sectional distributions of price stickiness that factor out sales improve the empirical fit of the model. Our results suggest that bridging the gap between micro and macro evidence on nominal price rigidity may require the combination of several mechanisms.

A random matrix theory approach to test for agricultural productivity convergence

Originating from multivariate statistics, random matrix theory (RMT) is used in order to test whether the elements of an empirical correlation coefficient matrix are noise dominated or contain true information. In this article, an attempt is made to apply the properties of RMT in macroeconomic time series data, by investigating the degree of convergence in agricultural labour productivity growth among a set of 32 European and Middle East and North Africa countries. Once the distribution of the eigenvalues of the empirical correlation matrix is found to differ from that of a pure random matrix, data are further analysed by means of hierarchical clustering techniques which allow for the creation of data clusters with common properties. This two-step procedure is an alternate means for club convergence tests, while some sensitivity analysis tests indicate an acceptable level of robustness of the proposed methodology even in small sample sizes.