A Nonlinear Approach for Modeling and Forecasting US Business Cycles

The purpose of this paper is to provide a complete evaluation of four regime-switching models by checking their performance in detecting US business cycle turning points, in replicating US business cycle features and in forecasting US GDP growth rate. Both individual and combined forecasts are considered. Results indicate that while the Markov-switching model succeeded in replicating all the NBER peak and trough dates without an extra-cycle detection, it seems to be outperformed by the Bounce-back model in term of the delay time to a correct alarm. Concerning business cycle features characterization, none of the competing models dominates over all the features. The performance of the Markov-switching and bounce back models in detecting turning points was not translated into an improved business cycle feature characterization since they are outperformed by the Floor and Ceiling model. The forecast performance of the considered models varies across regimes and across forecast horizons. That is, the model performing best in an expansion period is not necessarily the same in a recession period and similarly for the forecast horizons. Finally, combining such individual forecasts generally leads to increased forecast accuracy especially for h=1.     

Was the recent downturn in US real GDP predictable?

This article uses a small set of variables – real GDP, the inflation rate and the short-term interest rate – and a rich set of models – atheoretical (time series) and theoretical (structural), linear and nonlinear, as well as classical and Bayesian models – to consider whether we could have predicted the recent downturn of the US real GDP. Comparing the performance of the models to the benchmark random-walk model by root mean-square errors, the two structural (theoretical) models, especially the nonlinear model, perform well on average across all forecast horizons in our ex post, out-of-sample forecasts, although at specific forecast horizons certain nonlinear atheoretical models perform the best. The nonlinear theoretical model also dominates in our ex ante, out-of-sample forecast of the Great Recession, suggesting that developing forward-looking, microfounded, nonlinear, dynamic stochastic general equilibrium models of the economy may prove crucial in forecasting turning points.     

Forecasting US recessions with various risk factors and dynamic probit models

This paper extends probit recession forecasting models by incorporating various recession risk factors and using the advanced dynamic probit modeling approaches. The proposed risk factors include financial market expectations of a gloomy economic outlook, credit or liquidity risks in the general economy, the risks of negative wealth effects resulting from the bursting of asset price bubbles, and signs of deteriorating macroeconomic fundamentals. The model specifications include three different dynamic probit models and the standard static model. The out-of-sample analysis suggests that the four probit models with the proposed risk factors can generate more accurate forecasts for the duration of recessions than the conventional static models with only yield spread and equity price index as the predictors. Among the four probit models, the dynamic and dynamic autoregressive probit models outperform the static and autoregressive models in terms of predicting the recession duration. With respect to forecasting the business cycle turning points, the static probit model is as good as the dynamic probit models by being able to flag an early warning signal of a recession.     

The out-of-sample forecasting performance of nonlinear models of regional housing prices in the US

This article provides out-of-sample forecasts of linear and nonlinear models of US and four Census subregions’ housing prices. The forecasts include the traditional point forecasts, but also include interval and density forecasts, of the housing price distributions. The nonlinear smooth-transition autoregressive model outperforms the linear autoregressive model in point forecasts at longer horizons, but the linear autoregressive and nonlinear smooth-transition autoregressive models perform equally at short horizons. In addition, we generally do not find major differences in performance for the interval and density forecasts between the linear and nonlinear models. Finally, in a dynamic 25-step ex-ante and interval forecasting design, we, once again, do not find major differences between the linear and nonlinear models. In sum, we conclude that when forecasting regional housing prices in the United States, generally the additional costs associated with nonlinear forecasts outweigh the benefits for forecasts only a few months into the future.     

Comparing linear and nonlinear forecasts for stock returns

We compare the out-of-sample performance of monthly returns forecasts for two indices, namely the Dow Jones (DJ) and the Financial Times (FT) indices. A linear and a nonlinear artificial neural network (ANN) model are used to generate the out-of-sample competing forecasts for monthly returns. Stationary transformations of dividends and trading volume are considered as fundamental explanatory variables in the linear model and the input variables in the ANN model. The comparison of out-of-sample forecasts is done on the basis of forecast accuracy, using the Diebold and Mariano test [J. Bus. Econ. Stat. 13 (1995) 253.], and forecast encompassing, using the Clements and Hendry approach [J. Forecast. 5 (1998) 559.]. The results suggest that the out-of-sample ANN forecasts are significantly more accurate than linear forecasts of both indices. Furthermore, the ANN forecasts can explain the forecast errors of the linear model for both indices, while the linear model cannot explain the forecast errors of the ANN in either of the two indices. Overall, the results indicate that the inclusion of nonlinear terms in the relation between stock returns and fundamentals is important in out-of-sample forecasting. This conclusion is consistent with the view that the relation between stock returns and fundamentals is nonlinear.     

Forecasting Canadian time series with the New Keynesian model

Abstract.  This paper assesses the out-of-sample forecasting accuracy of the New Keynesian Model for Canada. We estimate a variant of the model on a series of rolling subsamples, computing out-of-sample forecasts one to eight quarters ahead at each step. We compare these forecasts with those arising from vector autoregression (VAR) models, using econometric tests of forecasting accuracy. We show that the forecasting accuracy of the New Keynesian Model compares favourably with that of the benchmarks, particularly as the forecasting horizon increases. These results suggest that the model could become a useful forecasting tool for Canadian time series.     

Bayesian forecasting of federal funds target rate decisions

In this paper we examine which macroeconomic and financial variables have most predictive ability for the federal funds target rate decisions made by the Federal Open Market Committee (FOMC). We conduct the analysis for the 157 FOMC decisions during the period January 1990–June 2008, using dynamic ordered probit models with a Bayesian endogenous variable selection methodology and real-time data for a set of 33 candidate predictor variables. We find that indicators of economic activity and forward-looking term structure variables, as well as survey measures are most informative from a forecasting perspective. For the full sample period, in-sample probability forecasts achieve a hit rate of 90%. Based on out-of-sample forecasts for the period January 2001–June 2008, 82% of the FOMC decisions are predicted correctly.     

Forecasting the semiconductor industry cycles by bootstrap prediction intervals

In recent years, there has been a recognition that point forecasts of the semiconductor industry sales may often be of limited value. There is substantial interest for a policy maker or an individual investor in knowing the degree of uncertainty that attaches to the point forecast before deciding whether to increase production of semiconductors or purchase a particular share from the semiconductor stock market. In this article, I first obtain the bootstrap prediction intervals of the global semiconductor industry cycles by a vector autoregressive (VAR) model using monthly US data consisting of four macroeconomic and seven industry-level variables with 92 observations. The 24-step-ahead out-of-sample forecasts from various VAR setups are used for comparison. The empirical result shows that the proposed 11-variable VAR model with the appropriate lag length captures the cyclical behaviour of the industry and outperforms other VAR models in terms of both point forecast and prediction interval.     

Forecasting stock market volatility: The role of technical variables

Accurate volatility forecasts are required by both market participants and policy makers. In this paper, we forecast stock return volatility by using a wide range of technical indicators constructed based on the past behavior of stock price, volatility and trading volume. Our out-of-sample results indicate that the incorporation of technical variables in the autoregression benchmark can produce significantly more accurate volatility forecasts. The forecasting performance of the combination of technical indicators is further compared with that of the popular economic indicators. Technical variables perform better than economic variables when the economy is an expansion, while the economic variables generate more accurate forecasts when the economy belongs a recession. These two types of variables provide complementary information over the business cycle. We obtain more reliable forecasts by combining all economic and technical information together than by combining either type of information alone.     

On business cycle fluctuations in USA macroeconomic time series

This study employs eighteen USA macroeconomic time series variables to investigate possible existence of asymmetries in business cycle fluctuations in the series. Detection of asymmetric fluctuations in economic activity is important for policymakers since effective monetary policy relies on asymmetric business cycle fluctuations in all the series. The asymmetric deviations from the long-term growth trend in each of the series are modeled using regime switching models and artificial neural networks. The results based on nonlinear switching time series models reveal strong evidence of business cycle asymmetries in most of the series. The results based on in-sample approximations from artificial neural networks show statistically significant evidence of asymmetries in all the series. Similar results are obtained when jackknife out-of-sample approximations from artificial neural networks are used. Thus, the study results show statistically significant evidence of asymmetries in all the series which indicates that business cycle fluctuations in the series are asymmetric, thus alike. Therefore, the impact of monetary policy shocks on the output and the other macroeconomic variables can be anticipated using nonlinear models only. The results on asymmetric business cycle fluctuations in real GDP are in line with recent studies but in sharp contrast with Balke and Fomby (1994).     

Forecasting the aggregate stock market volatility in a data-rich world

ABSTRACT

In this article, we utilize the basic lasso and elastic net models to revisit the predictive performance of aggregate stock market volatility in a data-rich world. Motivated by the existing literature, we determine several candidate predictors that have 22 technical indicators and 14 macroeconomic and financial variables. Our out-of-sample results reveal several noteworthy findings. First, few macroeconomic and financial variables and most of technical indicators have superior performance relative to the benchmark model. Second, combination forecasts are able to significantly beat the benchmark and some signal predictors Third, the lasso and elastic models with all predictors can generate more accurate forecasts than the benchmark and some other predictors in both the statistical and economic sense. Fourth, the lasso and elastic models exhibit higher forecast accuracy during periods of expansions and recessions. Finally, our findings are robust to several tests, such as different forecasting windows, forecasting models, and forecasting evaluations.     

On the uncertainty and risks of macroeconomic forecasts: combining judgements with sample and model information

Institutions which publish macroeconomic forecasts usually do not rely on a single econometric model to generate their forecasts. The combination of judgements with information from different models complicates the problem of characterizing the predictive density. This article proposes a parametric approach to construct the joint and marginal densities of macroeconomic forecasting errors, combining judgements with sample and model information. We assume that the relevant variables are linear combinations of latent independent two-piece normal variables. The baseline point forecasts are interpreted as the mode of the joint distribution, which has the convenient feature of being invariant to judgments on the balance of risks.     

Forecasting the yield curve and the role of macroeconomic information in Turkey

In this study we investigate the yield curve forecasting performance of Dynamic Nelson–Siegel Model (DNS), affine term structure VAR model (ATSM VAR) and principal component model (PC) in Turkey. We also investigate the role of macroeconomic variables in forecasting the yield curve. We have reached numbers of important results: 1—Macroeconomic variables are very useful in forecasting the yield curve. 2—The forecasting performances of the models depend on the period under review. 3—Considering the structural break which associates with change in monetary policy leads models to produce better forecasts than the random walk. 4—The role of exchange rate should not be ruled out in forecasting the yield curve in an emerging market like Turkey.     

Predicting Canadian recessions using dynamic probit modelling approaches

Abstract This paper examines the ability of various financial and macroeconomic variables to forecast Canadian recessions. It evaluates four model specifications, including the advanced dynamic, autoregressive, dynamic autoregressive probit models as well as the conventional static probit model. The empirical results highlight several significant recession predictors, notably the government bond yield spread, growth rates of the housing starts, the real money supply and the composite index of leading indicators. Both the in‐sample and out‐of‐sample results suggest that the forecasting performance of the four probit models is mixed. The dynamic and dynamic autoregressive probit models are better in predicting the duration of recessions while the static and autoregressive probit models are better in forecasting the peaks of business cycles. Hence, the advanced dynamic models and the conventional static probit model can complement one another to provide more accurate forecasts for the duration and turning points of business cycles.     

Macroeconomic and credit forecasts during the Greek crisis using Bayesian VARs

We investigate the ability of small- and medium-scale Bayesian VARs (BVARs) to produce accurate macroeconomic (output and inflation) and credit (loans and lending rate) out-of-sample forecasts during the latest Greek crisis. We implement recently proposed Bayesian shrinkage techniques based on Bayesian hierarchical modeling, and we evaluate the information content of forty-two (42) monthly macroeconomic and financial variables in terms of point and density forecasting. Alternative competing models employed in the study include Bayesian autoregressions (BARs) and time-varying parameter VARs with stochastic volatility, among others. The empirical results reveal that, overall, medium-scale BVARs enriched with economy-wide variables can considerably and consistently improve short-term inflation forecasts. The information content of financial variables, on the other hand, proves to be beneficial for the lending rate density forecasts across forecasting horizons. Both of the above-mentioned results are robust to alternative specification choices, while for the rest of the variables smaller-scale BVARs, or even univariate BARs, produce superior forecasts. Finally, we find that the popular, data-driven, shrinkage methods produce, on average, inferior forecasts compared to the theoretically grounded method considered here.     

Do we need a global VAR model to forecast inflation and output in South Africa?

This study determines whether the global vector autoregressive (GVAR) approach provides better forecasts of key South African variables than a vector error correction model (VECM) and a Bayesian vector autoregressive (BVAR) model augmented with foreign variables. The article considers both a small GVAR model and a large GVAR model in determining the most appropriate model for forecasting South African variables. We compare the recursive out-of-sample forecasts for South African GDP and inflation from six types of models: a general 33 country (large) GVAR, a customized small GVAR for South Africa, a VECM for South Africa with weakly exogenous foreign variables, a BVAR model, autoregressive (AR) models and random walk models. The results show that the forecast performance of the large GVAR is generally superior to the performance of the customized small GVAR for South Africa. The forecasts of both the GVAR models tend to be better than the forecasts of the augmented VECM, especially at longer forecast horizons. Importantly, however, on average, the BVAR model performs the best when it comes to forecasting output, while the AR(1) model outperforms all the other models in predicting inflation. We also conduct ex ante forecasts from the BVAR and AR(1) models over 2010:Q1–2013:Q4 to highlight their ability to track turning points in output and inflation, respectively.     

On the predictability of time-varying VAR and DSGE models

Over the last few years, there has been a growing interest in DSGE modelling for predicting macroeconomic fluctuations and conducting quantitative policy analysis. Hybrid DSGE models have become popular for dealing with some of the DSGE misspecifications as they are able to solve the trade-off between theoretical coherence and empirical fit. However, these models are still linear and they do not consider time variation for parameters. The time-varying properties in VAR or DSGE models capture the inherent nonlinearities and the adaptive underlying structure of the economy in a robust manner. In this article, we present a state-space time-varying parameter VAR model. Moreover, we focus on the DSGE–VAR that combines a microfounded DSGE model with the flexibility of a VAR framework. All the aforementioned models as well simple DSGEs and Bayesian VARs are used in a comparative investigation of their out-of-sample predictive performance regarding the US economy. The results indicate that while in general the classical VAR and BVARs provide with good forecasting results, in many cases the TVP–VAR and the DSGE–VAR outperform the other models.     

Ex ante forecasts of business-cycle turning points

This paper presents a logit model that accurately forecasts business-cycle turning points with a lead of one-quarter. The sample period consists of an initialization subset (1959:Q3–1975:Q4), and a subset for out-of-sample forecast evaluation (1976:Q1–2005:Q4). In contrast with the record of disappointing results in the literature, the model correctly forecasts all turning points in the test subset without forecasting recessions that did not occur.      

Does non-linearity help us understand,model and forecast UK stock and bond returns: evidence from the BEYR

The usefulness of non-linear models to provide accurate estimates and forecasts remains an open empirical debate. This paper examines the nature of the estimated relationships and forecasting power of smooth-transition models for UK stock and bond returns using a range of financial and macroeconomic variables as predictors. Notably, evidence of non-linearity is stronger when the bond-equity yield ratio is used as the transition variable. This ratio measures whether stocks are over (under)-valued relative to bonds and can act as a signal for portfolio managers. In-sample results reveal noticeable differences regarding the nature of relationships between the linear and non-linear setting, while results of a recursive forecasting exercise reveal both statistical and economic improvement over a linear model. Overall, these results support the view that non-linear estimates and forecasts can provide useful information for stock market traders, portfolio managers and policy-makers.     

Sentiment dynamics and stock returns: the case of the German stock market

We use weekly survey data on short-term and medium-term sentiment of German investors in order to study the causal relationship between investors’ mood and subsequent stock price changes. In contrast to extant literature for other countries, a trivariate vector autoregression for short-run sentiment, medium-run sentiment, and stock index returns allows to reject exogeneity of returns. Depending on the chosen VAR specification, returns are found to either follow a feedback process caused by medium-run sentiment, or returns form a simultaneous systems together with the two sentiment measures. An out-of-sample forecasting experiment on the base of estimated subset VAR models shows significant exploitable linear structure. However, trading experiments do not yield convincing evidence of significant economic gains from the VAR forecasts, and it appears that predictability of returns from sentiment decreases during the recent market gyrations.