FORECASTING PERFORMANCE OF AN ESTIMATED DSGE MODEL FOR THE SOUTH AFRICAN ECONOMY

We construct a small open‐economy New Keynesian dynamic stochastic general equilibrium (DSGE) model for South Africa with nominal rigidities, incomplete international risk sharing and partial exchange rate pass‐through. The parameters of the model are estimated using Bayesian methods, and its out‐of‐sample forecasting performance is compared with Bayesian vector autoregression (VAR), classical VAR and random‐walk models. Our results indicate that the DSGE model generates forecasts that are competitive with those from other models, and it contributes statistically significant information to combined forecast measures.     

SPATIAL BAYESIAN METHODS OF FORECASTING HOUSE PRICES IN SIX METROPOLITAN AREAS OF SOUTH AFRICA

This paper estimates Spatial Bayesian Vector Autoregressive (SBVAR) models, based on the First-Order Spatial Contiguity and the Random Walk Averaging priors, for six metropolitan areas of South Africa, using monthly data over the period of 1993:07 to 2005:06. We then forecast one- to six-months-ahead house prices over the forecast horizon of 2005:07 to 2007:06. When we compare forecasts generated from the SBVARs with those from an unrestricted Vector Autoregressive (VAR) and the Bayesian Vector Autoregressive (BVAR) models based on the Minnesota prior, we find that the spatial models tend to outperform the other models for large middle-segment houses; while the VAR and the BVAR models tend to produce lower average out-of-sample forecast errors for middle and small-middle segment houses, respectively. In addition, based on the priors used to estimate the Bayesian models, our results also suggest that prices tend to converge for both large- and middle-sized houses, but no such evidence could be obtained for the small-sized houses.     

FORECASTING THE SOUTH AFRICAN ECONOMY WITH VARs AND VECMs

The paper develops a Bayesian Vector Error Correction Model (BVECM) of the South African economy for the period 1970:1‐2000:4 and forecasts GDP, consumption, investment, short and long term interest rates, and the CPI. We find that a tight prior produces relatively more accurate forecasts than a loose one. The out‐of‐sample‐forecast accuracy resulting from the BVECM is compared with those generated from the Classical variant of the VAR and VECM and the Bayesian VAR. The BVECM is found to produce the most accurate out of sample forecasts. It also correctly predicts the direction of change in the chosen macroeconomic indicators.     

A BVAR MODEL FOR THE SOUTH AFRICAN ECONOMY

The paper develops a Bayesian vector autoregressive (BVAR) model of the South African economy for the period of 1970:1‐2000:4 and forecasts GDP, consumption, investment, short‐term and long term interest rates, and the CPI. We find that a tight prior produces relatively more accurate forecasts than a loose one. The out‐of‐sample‐forecast accuracy resulting from the BVAR model is compared with the same generated from the univariate and unrestricted VAR models. The BVAR model is found to produce the most accurate out of sample forecasts. The same is also capable of correctly predicting the direction of change in the chosen macroeconomic indicators.     

Learning about fundamentals: The widening of the French ERM bands in 1993

Learning about Fundamentals: The Widening of the French ERM Bands in 1993. — The authors incorporate a Bayesian learning model into a fairly general model of exchange rate determination in discrete time. The model is applied to the period following the widening of the French-German ERM bands in August 1993, in which a systematic underprediction of the franc can be observed until February 1994. A (substantial) part of these forecast errors can be mimicked by a Bayesian learning process. Simulations with our model show that, after the widening of the bands, agents, contrary to their initial expectations, gradually learned that the true process driving monetary conditions in France had not changed notably.     

BAYESIAN METHODS OF FORECASTING INVENTORY INVESTMENT

This paper develops a Bayesian Vector Error Correction Model (BVECM) for forecasting inventory investment. The model is estimated using South African quarterly data on actual sales, production, unfilled orders, price level and interest rate, for the period 1978 to 2000. The out-of-sample-forecast accuracy obtained from the BVECM over the forecasting horizon of 2001:1 to 2003:4, is compared with those generated from the classical variant of the Vector Autoregresssive (VAR) model and the VECM, the Bayesian VAR, and the recently developed ECM by Smith et al. , for the South African economy. The BVECM with the most-tight prior outperforms all the other models, except for a relatively tight BVAR which also correctly predicts the direction of change of inventory investment over the period of 2004:1 to 2006:3.     

FORECASTING THE SOUTH AFRICAN ECONOMY WITH GIBBS SAMPLED BVECMs

The paper uses the Gibbs sampling technique to estimate a heteroscedastic Bayesian Vector Error Correction Model (BVECM) of the South African economy for the period 1970:1‐2000:4, and then forecasts GDP, consumption, investment, short and long term interest rates, and the CPI over the period of 2001:1 to 2005:4. We find that a tight prior produces relatively more accurate forecasts than a loose one. The out‐of‐sample‐forecast accuracy resulting from the Gibbs sampled BVECM is compared with those generated from a Classical VECM and a homoscedastic BVECM. The homoscedastic BVECM is found to produce the most accurate out of sample forecasts.     

Inflation Forecast Targeting: An Alternative Approach to Estimating the Inflation‐Output Variability Tradeoff

We suggest a new way of computing the inflation‐output variability tradeoff under inflation forecast targeting. Our approach is based on dynamic, stochastic simulations of the average inflation rate over a two‐year horizon using the moving average representation of a vector autoregressive (VAR) model. Using real‐time data over two samples, we estimate the inflation‐output variability tradeoff for the United States and show that it has shifted favorably over time. We analyze the policy interventions required to achieve target inflation in each sample and compare these interventions over time.     

Interdependence between The Netherlands and Germany: Forecasting with VAR models

Summary The German economy is usually assumed to take a leading position. In principle this gives smaller countries, which are dependent on Germany, the opportunity to predict their own economic future conditional on the state of the German economy. This paper uses this opportunity for The Netherlands by applying a Vector Auto Regressive model on Dutch and German series. Because the traditional VAR models appear to be overparameterized, their forecast performance can be improved significantly by using shrinkage estimators based on the so-called Minnesota prior. Such a Bayesian VAR forecasts well and confirms the interdependence between Germany and The Netherlands. Variance decomposition of forecast errors and impulse response simulations strengthen the impression that the BVAR model properties are plausible.The author works at the Econometric Research and Special Studies Department of the Nederlandsche Bank. He is grateful to M.M.G. Fase, C.C.A. Winder and two anonymous referees for their useful comments and to R.B.M. Vet for his assistance in various calculations.     

An analysis of RSQE forecasts: 1971–1992

The purpose of this paper is to evaluate the accuracy of ex ante econometric model forecasts of four key macroeconomic variables: real GNP growth, the rate of price inflation measured by the GNP deflator, the civilian unemployment rate, and the Treasury Bill rate. Annual forecasts produced by the Research Seminar in Quantitative Economics (RSQE) based on the Michigan Quarterly Econometric Model of the U.S. Economy are compared with quasi ex ante forecasts from a four-variable vector autoregressive (VAR) model. Statistical tests of the equality of forecast error variances as well as univariate and multivariate forecast encompassing-type tests are conducted. The forecast error variance comparisons indicate that for three of the four variables the RSQE forecasts are more accurate than the VAR forecasts and for one of the variables (real GNP growth) only slightly less accurate. The forecast encompassing-type tests indicate that the RSQE forecasts contain information not contained in the VAR forecasts and, conversely, that VAR forecasts contain information not included in the RSQE forecasts. The scope for improving RSQE forecasts by combining them with VAR forecasts is rather limited, however.     

On the choice of monetary policy rules for China: A Bayesian DSGE approach

Motivated by the institutional features of China's monetary policy, this paper aims at identifying the most data favored monetary policy rule for China within a dynamic stochastic general equilibrium (DSGE) model framework. In a canonical New-Keynesian DSGE model, we carry out a positive analysis by employing Bayesian methods to estimate three main categories of monetary policy rules, namely a Taylor-type interest rate rule, a money growth rule and an expanded Taylor rule with money. Based on China's quarterly data from 1996Q2 to 2015Q4, our estimation shows that the expanded Taylor rule obtains the best empirical fit to the data. Moreover, impulse responses and forecast error variance decompositions demonstrate that monetary policy rules with or without money provide very different implications for the policy behavior. Our results ultimately suggest that money has so far been more closely targeted than nominal interest rate and still plays an important role as a monetary policy target in China. Furthermore, a conventional Taylor-type interest rate rule is not good enough yet to describe China's monetary policy behavior.     

Occupational Mobility Pattern in Chonju,Korea

This paper investigates the existence of the inter‐dependence between the Indian stock market and Asia's emerging markets since 1990. This study analyzes whether the MSCI Asian Index has significantly influenced the Bombay Stock Exchange Index before, during, and after the Asian financial crisis. To address this issue, the author first uses a rolling correlation, and conduct uni‐directional and bi‐directional causality tests using the Granger causality test. He then examines the impulse response functions and variance decompositions of forecast errors based on a VAR (vector auto‐regression) model. These tests provide evidence that the influence of the Asian market on the Indian market has increased during and after the Asian financial crisis. These results can be interpreted as evidence that the Indian market has been moving toward integration with other Asian markets.     

What Drives Ireland’s Housing Market? A Bayesian DSGE Approach

We study the drivers of fluctuations in the Irish housing market by developing and estimating a dynamic stochastic general equilibrium (DSGE) model of Ireland as a member of the European Economic Monetary Union (EMU). We estimate the model with Bayesian methods using time series for both Ireland and the rest of the EMU for the period from 1997:Q1 to 2008:Q2. We find that housing preference and technology shocks are the main drivers of fluctuations in house prices and residential investment. A standard regression analysis shows that a good part of the variation of housing preference shocks can be explained by unmodeled demand factors that have been considered in the empirical literature as important determinants of Irish house prices.     

IMPLEMENTING A ROBUST RISK MODEL FOR SOUTH AFRICAN EQUITY MARKETS: A PEAK‐OVER‐THRESHOLD APPROACH

This paper implements a market risk model for the South African equity market using daily returns of the Johannesburg Stock Exchange All Share Index. Firstly, we separate positive returns from negative returns and model them using the peak‐over‐threshold (POT) method in order to compute the downside as well as upside risk measures separately. We thereafter compute the value‐at‐risk (VAR) and the expected shortfall (ES) estimates corresponding to upside and downside risks. We bootstrap these risk measures and compute their standard errors and confidence intervals (CIs) to see whether they fall inside these CIs. Secondly, we compute out‐sample forecasts of VAR estimates using the POT method and the generalised autogressive conditional heteroscedasticity process. Three backtesting methodologies are employed: the unconditional and conditional coverage tests and the counting of number of exceptions according to Basel II green zone. We find that all our VAR and ES estimates are well inside their CIs and that at lower quantiles, parametric ES estimates are equal to POT‐ES estimates, although the difference between the two is more pronounced at higher quantiles (99% or higher). Furthermore, our market risk model falls into the Basel II green zone, as it produces fewer exceptions in out‐sample space.     

Bayesian analysis of time-varying parameter vector autoregressive model for the Japanese economy and monetary policy

This paper analyzes the time-varying parameter vector autoregressive (TVP–VAR) model for the Japanese economy and monetary policy. The parameters are allowed to follow a random walk process and estimated using the Markov chain Monte Carlo method. The empirical result reveals the time-varying structure of the Japanese economy and monetary policy during the period from 1981 to 2008. The marginal likelihoods of the TVP–VAR model and other fixed parameter VAR models are estimated for model comparison. The estimated marginal likelihoods indicate that the TVP–VAR model best fits the Japanese economic data.     

How Useful are DSGE Macroeconomic Models for Forecasting?

A review of the literature shows that forecasts from DSGE models are not more accurate than either times series models or official forecasts, but neither are they any worse. Further, all three types of forecast failed to predict the recession that started in 2007 and continued to forecast poorly even after the recession was known to have begun. The aim of this paper is to investigate why these results occur by examining the structure of the solution of DSGE models and compare this with pure time series models. The main factor seems to be the dynamic structure of DSGE models. Their backward-looking dynamics gives them a similar forecasting structure to time series models and their forward-looking dynamics, which consists of expected values of future exogenous variables, is difficult to forecast accurately. This suggests that DSGE models should not be tested through their forecasting ability.     

Monetary policy and inflation in South Africa: A VECM augmented with foreign variables

We develop a structural cointegrated vector autoregressive (VAR) model with weakly exogenous foreign variables, known as an augmented VECM or VECX*, suitable for a small open economy like South Africa. This model is novel for South Africa in two ways: it is the first VECX* developed to analyse monetary policy and the first model that uses time‐varying trade weights to create the foreign series. We impose three significant long‐run relations (augmented purchasing power parity, uncovered interest parity and Fisher parity) to investigate the effect of a monetary policy shock on inflation. The results suggest the effective transmission of monetary policy.     

Auditor Quality and the Accuracy of Management Earnings Forecasts*

In this study, we appeal to insights and results from Davidson and Neu 1993 and McConomy 1998 to motivate empirical analyses designed to gain a better understanding of the relationship between auditor quality and forecast accuracy. We extend and refine Davidson and Neu's analysis of this relationship by introducing additional controls for business risk and by considering data from two distinct time periods: one in which the audit firm's responsibility respecting the earnings forecast was to provide review‐level assurance, and one in which its responsibility was to provide audit‐level assurance. Our sample data consist of Toronto Stock Exchange (TSE) initial public offerings (IPOs). The earnings forecast we consider is the one‐year‐ahead management earnings forecast included in the IPO offering prospectus. The results suggest that after the additional controls for business risk are introduced, the relationship between forecast accuracy and auditor quality for the review‐level assurance period is no longer significant. The results also indicate that the shift in regimes alters the fundamental nature of the relationship. Using data from the audit‐level assurance regime, we find a negative and significant relationship between forecast accuracy and auditor quality (i.e., we find Big 6 auditors to be associated with smaller absolute forecast errors than non‐Big 6 auditors), and further, that the difference in the relationship between the two regimes is statistically significant.     

DYNAMIC TIME INCONSISTENCY AND THE SOUTH AFRICAN RESERVE BANK

This paper derives the econometric restrictions imposed by the Barro and Gordon model of dynamic time inconsistency on a bivariate time-series model of consumer price index (CPI) inflation and real gross domestic product (GDP), and tests these restrictions based on quarterly data for South Africa covering the period of January 1960-April 1999, i.e. for the pre-inflation targeting period. The results show that the data are consistent with the short- and long-run implications of the theory of time-consistent monetary policy. Moreover, when the model is used to forecast one-step-ahead inflation over the period of January 2001-February 2008, i.e. the period covering the starting point of the inflation-targeting regime until date, we, on average, obtain lower rates of inflation. The result tends to suggest that the South African Reserve Bank perhaps needs to manage the inflation-targeting framework better than it has done so far.     

Economic Fluctuations in the U.S. and Euro Area: Quantifying the Contribution of Technical Change

This article aims at quantifying the contribution of technical change to cyclical fluctuations in the U.S. and euro area. We distinguish technical progress in labor-augmenting and capital-augmenting change. To this end, we derive and estimate a New Keynesian DSGE model embodying a constant elasticity of substitution (CES) production function for both areas. Our main findings are: (i) capital-augmenting progress is the main source of technical change volatility; (ii) labor-augmenting shocks give a negligible contribution to the variance of output; (iii) technical change (of both types) explains more economic fluctuations in the U.S. than in the euro area; and (iv) historical decomposition of GDP growth over our sample period (1980–2008) shows that capital-augmenting progress is one of the key drivers of the business cycle.