Search results

Recently, it has been suggested that macroeconomic forecasts from estimated dynamic stochastic general equilibrium (DSGE) models tend to be more accurate out-of-sample than random walk forecasts or Bayesian vector autoregression (VAR) forecasts. Del Negro and Schorfheide (2013) in particular suggest that the DSGE model forecast should become the benchmark for forecasting horse-races. We compare the real-time forecasting accuracy of the Smets and Wouters (2007) DSGE model with that of several reduced-form time series models. We first demonstrate that none of the forecasting models is efficient. Our second finding is that there is no single best forecasting method. For example, typically simple AR models are most accurate at short horizons and DSGE models are most accurate at long horizons when forecasting output growth, while for inflation forecasts the results are reversed. Moreover, the relative accuracy of all models tends to evolve over time. Third, we show that there is no support to the common practice of using large-scale Bayesian VAR models as the forecast benchmark when evaluating DSGE models. Indeed, low-dimensional unrestricted AR and VAR forecasts may forecast more accurately.

Analyses and checks the annual forecasts produced each autumn from four prominent UK economic modelling organizations. Compares these forecasts with those of three Bayesian vector‐autoregressive models. Examines the accuracy for each set of forecasts up to four years ahead and for different horizons. Examines the direction of the forecasts and the effect of forming simple combinations of the different forecasts. Finds evidence that while the BVAR forecasts are inferior to those from the economic models, they contain information which could be used in order to improve the other forecasts.

Small-scale VARs are widely used in macroeconomics for forecasting US output, prices, and interest rates. However, recent work suggests these models may exhibit instabilities. As such, a variety of estimation or forecasting methods might be used to improve their forecast accuracy. These include using different observation windows for estimation, intercept correction, time-varying parameters, break dating, Bayesian shrinkage, model averaging, etc. This paper compares the effectiveness of such methods in real-time forecasting. We use forecasts from univariate time series models, the Survey of Professional Forecasters, and the Federal Reserve Board's Greenbook as benchmarks.

This article surveys recent developments in the evaluation of point and density forecasts in the context of forecasts made by vector autoregressions. Specific emphasis is placed on highlighting those parts of the existing literature that are applicable to direct multistep forecasts and those parts that are applicable to iterated multistep forecasts. This literature includes advancements in the evaluation of forecasts in population (based on true, unknown model coefficients) and the evaluation of forecasts in the finite sample (based on estimated model coefficients). The article then examines in Monte Carlo experiments the finite-sample properties of some tests of equal forecast accuracy, focusing on the comparison of VAR forecasts to AR forecasts. These experiments show the tests to behave as should be expected given the theory. For example, using critical values obtained by bootstrap methods, tests of equal accuracy in population have empirical size about equal to nominal size.

This chapter investigates the performance of different dimension reduction approaches for large vector autoregressions in multi-step ahead forecasts. The authors consider factor augmented VAR models using principal components and partial least squares, random subset regression, random projection, random compression, and estimation via LASSO and Bayesian VAR. The authors compare the accuracy of iterated and direct multi-step point and density forecasts. The comparison is based on macroeconomic and financial variables from the FRED-MD data base. Our findings suggest that random subspace methods and LASSO estimation deliver the most precise forecasts.

This study aims to test the validity of the Rajan theory in South Africa and other selected emerging markets (Chile, Peru and Brazil) during the period 1975–2019.

In this study, the researchers used time-series data to estimate a Bayesian Vector Autoregression (BVAR) model with hierarchical priors. The BVAR technique has the advantage of being able to accommodate a wide cross-section of variables without running out of degrees of freedom. It is also able to deal with dense parameterization by imposing structure on model coefficients via prior information and optimal choice of the degree of formativeness.

The results for all countries except Peru confirmed the Rajan hypotheses, indicating that inequality contributes to high indebtedness, resulting in financial fragility. However, for Peru, this study finds it contradicts the theory. This study controlled for monetary policy shock and found the results differing country-specific.

The findings suggest that an escalating level of inequality leads to financial fragility, which implies that policymakers ought to be cautious of excessive inequality when endeavouring to contain the risk of financial fragility, by implementing sound structural reform policies that aim to attract investments consistent with job creation, development and growth in these countries. Policymakers should also be cautious when implementing policy tools (redistributive policies, a sound monetary policy), as they seem to increase the risk of excessive credit growth and financial fragility, and they need to treat income inequality as an important factor relevant to macroeconomic aggregates and financial fragility.

We develop a Bayesian approach for modeling brain activation and connectivity from functional magnetic resonance image (fMRI) data. Our approach simultaneously estimates local hemodynamic response functions (HRFs) and activation parameters, as well as global effective and functional connectivity parameters. Existing methods assume identical HRFs across brain regions, which may lead to erroneous conclusions in inferring activation and connectivity patterns. Our approach addresses this limitation by estimating region-specific HRFs. Additionally, it enables neuroscientists to compare effective connectivity networks for different experimental conditions. Furthermore, the use of spike and slab priors on the connectivity parameters allows us to directly select significant effective connectivities in a given network.

We include a simulation study that demonstrates that, compared to the standard generalized linear model (GLM) approach, our model generally has higher power and lower type I error and bias than the GLM approach, and it also has the ability to capture condition-specific connectivities. We applied our approach to a dataset from a stroke study and found different effective connectivity patterns for task and rest conditions in certain brain regions of interest (ROIs).

The study examines the dynamic relationship among globalisation, income inequality and poverty in Mexico, Indonesia, Nigeria and Turkey (MINT countries) between 1980 and 2018.

A Bayesian vector autoregressive (BVAR) approach is used as a technique of estimation hanging on the fact that the method uses prior distribution for the estimated parameters.

The results show that globalisation is a strong predictor of itself in all the MINT countries only in the short run. In the long run, income inequality and poverty strongly influence globalisation, respectively, in Indonesia and Turkey, while globalisation still has more impact on itself in Nigeria. Income inequality has a strong endogenous impact on itself in Mexico and Indonesia over the time horizon, whereas globalisation and poverty are strong predictors of income inequality in the long run in Nigeria and Turkey, respectively. Also, poverty strongly influences itself in all the MINT countries in all the periods, meaning that poverty begets itself in all the MINT countries, except for Indonesia in the long run.

The study suggests that all the MINT countries should ensure political stability and a strong institutional framework to gain from the process of globalisation and to experience reductions in the levels of income inequality and poverty.

This study is distinct from other studies in the sense that an overall globalisation index (GBI) as used by Dreher et al. (2008) is used for the globalisation variable, and the Multidimensional Poverty Index (MPI) is used to capture poverty in all the MINT countries. Also, the research paper uses a BVAR approach as against the classical VAR, and this helps in solving over-fitting problems.

Examines the causal relationship between interest rates, savings and income in the Chinese economy over the period 1952 to 1999, using the cointegration test and Bayesian vector autoregression (BVAR) for empirical testing. The empirical evidence from the cointegration test confirms that there is a stable long‐run relationship between interest rates, savings and income, whilst the BVAR causality test shows unidirectional causality running from savings to income. For China’s transitional economy, it is therefore important to establish well‐developed financial institutions – particularly the independence of the Central Bank – interest rate liberalization and sound financial intermediation, all of which are important for the efficient allocation of capital, which, in turn, can help to establish sustainable economic growth.

The objective of this paper is to explore the sensitivity of industry‐specific stock returns to monetary policy and macroeconomic news. The paper looks at a range of industry‐specific South African stock market indices and evaluates the sensitivity of these indices to various unanticipated macroeconomic shocks.

The authors begin with an event study, which examines the immediate impact of macroeconomic shocks on the stock market indices, and then use a Bayesian vector autoregressive (BVAR) analysis, which provides insight into the dynamic effects of the shocks on the stock market indices, by allowing them to treat the shocks as exogenous through appropriate setting of priors defining the mean and variance of the parameters in the VAR.

The results from the event study indicate that with the exception of the gold mining index, where the CPI surprise plays a significant role, monetary surprise is the only variable that consistently negatively affects the stock returns significantly, both at the aggregate and sectoral levels. The BVAR model based on monthly data, however, indicates that, in addition to the monetary policy surprises, the CPI and PPI surprises also affect aggregate stock returns significantly. However, the effects of the CPI and PPI surprises are quite small in magnitude and are mainly experienced at shorter horizons immediately after the shock.

To the best of the authors' knowledge, this is the first study conducted on South Africa which analyses the impact of a wide range of unanticipated macroeconomic shocks on stock returns. This paper improves on earlier efforts by using measures of monetary policy, as well as other macroeconomic news, which more cleanly isolates the unanticipated elements of the monetary policy variable and other macroeconomic indicators, in studying the impact of these surprises on stock returns in South Africa.