Monetary policy shocks and financial conditions: A Monte Carlo experiment

The effects of monetary policy shocks on financial conditions are often estimated by appealing to recursive Vector AutoRegressions (VARs). We assess the ability of this class of VARs to recover the true effects of a monetary policy shock via a Monte Carlo experiment in which the Data Generating Process is a Dynamic Stochastic General Equilibrium (DSGE) model featuring macro-finance interactions and estimated with U.S. quarterly data. Our DSGE model predicts a negative and significant reaction of financial conditions to an unexpected monetary policy tightening. We point out that such reaction is just overlooked by recursive VARs. Moreover, we show that Cholesky-VARs may substantially underestimate the welfare costs due to macroeconomic fluctuations.     

Putting the New Keynesian DSGE Model to the Real‐Time Forecasting Test

The paper compares the quality of real‐time forecasts from a standard medium‐scale New Keynesian dynamic stochastic general equilibrium (DSGE) model to those from the Survey of Professional Forecasters (SPF) and DSGE‐VARs. It is shown that the DSGE model is relatively successful in forecasting the U.S. economy. This is especially true for forecasts conditional on SPF nowcasts, in which case the forecasting power of the DSGE turns out to be similar or better than that of the SPF for all the variables and horizons. An important weakness of the benchmark DSGE model is the poor absolute performance of its point forecasts and rather badly calibrated forecast densities.     

Predicting Long‐Term Financial Returns: VAR versus DSGE Model—A Horse Race

This paper considers a U.S. institutional investor who is implementing a long‐term portfolio allocation using forecasts of financial returns. We compare the predictive performance of two competing macrofinance models—an unrestricted vector autoRegression (VAR) and a fully‐structural dynamic stochastic general equilibrium (DSGE) model—for horizons up to 15 years. Although the performances are similar for short horizons, the DSGE model outperforms the VAR at forecasting financial returns in the long term. This model also generates substantially higher Sharpe ratios. Although it contains fewer unknown parameters, it benefits from economically grounded restrictions that help anchor financial returns in the long term.     

Vector autoregressions and reduced form representations of DSGE models

The performance of dynamic stochastic general equilibrium models is often tested against estimated VARs. This requires that the data-generating process consistent with the DSGE theoretical model has a finite order VAR representation. This paper discusses the assumptions needed for a finite order VAR(p) representation of a DSGE model to exist. When a VAR(p) is only an approximation to the exact infinite order VAR, the truncated VAR(p) may return largely incorrect estimates of the impulse response function. The results do not hinge on small-sample bias or on incorrect identification assumptions. But the bias introduced by truncation can lead to bias in the identification of the structural shocks. Identification strategies that work in the exact VAR representation perform poorly in the truncated VAR.     

Uncertainty shocks and unemployment dynamics in U.S. recessions

What are the effects of uncertainty shocks on unemployment dynamics? We answer this question by estimating non-linear (Smooth-Transition) VARs with post-WWII U.S. data. The relevance of uncertainty shocks is found to be much larger than that predicted by standard linear VARs in terms of (i) magnitude of the reaction of the unemployment rate to such shocks, and (ii) contribution to the variance of the prediction errors of unemployment at business cycle frequencies. The ability of different classes of DSGE models to replicate our results is discussed.     

Futures volatility forecasting based on big data analytics with incorporating an order imbalance effect

Future markets play vital roles in supporting economic activities in modern society. For example, crude oil and electricity futures markets have heavy effects on a nation’s energy operation management. Thus, volatility forecasting of the futures market is an emerging but increasingly influential field of financial research. In this paper, we adopt big data analytics, called Extreme Gradient Boosting (XGBoost) from computer science, in an attempt to improve the forecasting accuracy of futures volatility and to demonstrate the application of big data analytics in the financial spectrum in terms of volatility forecasting. We further unveil that order imbalance estimation might incorporate abundant information to reflect price jumps and other trading information in the futures market. Including order imbalance information helps our model capture underpinned market rules such as supply and demand, which lightens the information loss during the model formation. Our empirical results suggest that the volatility forecasting accuracy of the XGBoost method considerably beats the GARCH-jump and HAR-jump models in both crude oil futures market and electricity futures market. Our results could also produce plentiful research implications for both policy makers and energy futures market participants.     

Tax Evasion,Investment Shocks,and the Consumption Puzzle: A DSGE Analysis with Financial Frictions

Recent studies identify Marginal Efficiency of Investment (MEI) shocks as important drivers of the business cycle. However, Dynamic Stochastic General Equilibrium (DSGE) models struggle to explain macroeconomic comovements between consumption and the key real variables after a MEI shock. Moreover, engaging in tax evasion practices is often an answer to financial constraints, which have been recognized as important determinants of cyclical fluctuations as well. We use a medium-scale New Keynesian DSGE model, that combines tax evasion with financial frictions, to simulate a MEI shock. We show that entrepreneurial tax evasion can solve the comovement problem to a fair extent.     

Indeterminacy and Forecastability

Recent studies document the deteriorating performance of forecasting models during the Great Moderation, which conversely implies that forecastability was higher in the preceding era when the economy was unexpectedly volatile. We explain this phenomenon in the context of equilibrium indeterminacy in dynamic stochastic general equilibrium (DSGE) models. We first analytically show that a model under indeterminacy exhibits richer dynamics that can improve forecastability. Then, using a sticky‐price DSGE model, we numerically demonstrate that indeterminacy arising from passive monetary policy generates persistent dynamics that lead to superior forecastability. We also point out the possibility that forecastability under indeterminacy deteriorates when the degree of uncertainty about sunspot fluctuations is large.     

Bankruptcy Prediction with Industry Effects

This paper investigates the forecasting accuracy of bankruptcy hazard rate models for U.S. companies over the time period 1962–1999 using both yearly and monthly observation intervals. The contribution of this paper is multiple-fold. One, using an expanded bankruptcy database we validate the superior forecasting performance of Shumway's (2001) model as opposed to Altman (1968) and Zmijewski (1984). Two, we demonstrate the importance of including industry effects in hazard rate estimation. Industry groupings are shown to significantly affect both the intercept and slope coefficients in the forecasting equations. Three, we extend the hazard rate model to apply to financial firms and monthly observation intervals. Due to data limitations, most of the existing literature employs only yearly observations. We show that bankruptcy prediction is markedly improved using monthly observation intervals. Fourth, consistent with the notion of market efficiency with respect to publicly available information, we demonstrate that accounting variables add little predictive power when market variables are already included in the bankruptcy model.     

Exploring international differences in inflation dynamics

Standard closed-economy DSGE models have difficulty replicating the persistence of inflation. We use a multicountry dataset to establish some empirical regularities on persistence and volatility of aggregate consumer prices for 135 countries since 1993. We find both persistence and volatility to be low (high) in developed (developing) countries relative to the full sample average. This pattern is also observed in low (high) inflation countries. We then employ a two-country DSGE framework to investigate the extent to which structural open economy features, such as incomplete exchange rate pass-through, the existence of nontraded goods, and international financial market incompleteness, can help in replicating the persistence and volatility of consumer prices. Our simulation results indicate that the model can replicate the degree of inflation persistence seen in the data for both developed and developing countries, but cannot generate the low levels of volatility observed in developed economies.     

Hierarchical GARCH

There is strong empirical evidence that the GARCH estimates obtained from panels of financial time series cluster. In order to capture this empirical regularity, this paper introduces the Hierarchical GARCH (HG) model. The HG is a nonlinear panel specification in which the coefficients of each series are modeled as a function of observed series characteristic and an unobserved random effect. A joint panel estimation strategy is proposed to carry out inference for the model. A simulation study shows that when there is a strong degree of coefficient clustering panel estimation leads to substantial accuracy gains in comparison to estimating each GARCH individually. The HG is applied to a panel of U.S. financial institutions in the 2007–2009 crisis, using firm size and leverage as characteristics. Results show evidence of coefficient clustering and that the characteristics capture a significant portion of cross sectional heterogeneity. An out-of-sample volatility forecasting application shows that when the sample size is modest coefficient estimates based on the panel estimation approach perform better than the ones based on individual estimation.     

Finance and international business cycles

Business cycles are more correlated among countries that have similar financial structures. We first document this empirical regularity using OECD data, and then build a two-country DSGE model with financial frictions that replicates it. Alternative monetary policy regimes and parameter values are explored. Output co-movements increase when the countries involved are linked by a credible exchange rate peg and when they open up to trade; they decrease when their financial openness increases. The model also accounts for a number of stylized facts of international business cycles, such as the positive international correlation of output, investment and employment.     

Predicting the long-term stock market volatility: A GARCH-MIDAS model with variable selection

We consider a GARCH-MIDAS model with short-term and long-term volatility components, in which the long-term volatility component depends on many macroeconomic and financial variables. We select the variables that exhibit the strongest effects on the long-term stock market volatility via maximizing the penalized log-likelihood function with an Adaptive-Lasso penalty. The GARCH-MIDAS model with variable selection enables us to incorporate many variables in a single model without estimating a large number of parameters. In the empirical analysis, three variables (namely, housing starts, default spread and realized volatility) are selected from a large set of macroeconomic and financial variables. The recursive out-of-sample forecasting evaluation shows that variable selection significantly improves the predictive ability of the GARCH-MIDAS model for the long-term stock market volatility.     

International stock return predictability: Evidence from new statistical tests

We investigate whether stock returns of international markets are predictable from a range of fundamentals including key financial ratios (dividend-price ratio, dividend-yield, earnings-price ratio, dividend-payout ratio), technical indicators (price pressure, change in volume), and short-term interest rates. We adopt two new alternative testing and estimation methods: the improved augmented regression method and wild bootstrapping of predictive model based on a restricted VAR form. Both methods take explicit account of endogeneity of predictors, providing bias-reduced estimation and improved statistical inference in small samples. From monthly data of 16 Asia-Pacific (including U.S.) and 21 European stock markets from 2000 to 2014, we find that the financial ratios show weak predictive ability with small effect sizes and poor out-of-sample forecasting performances. In contrast, the price pressure and interest rate are found to be strong predictors for stock return with large effect sizes and satisfactory out-of-sample forecasting performance.     

Semi-parametric Bayesian tail risk forecasting incorporating realized measures of volatility

Realized measures employing intra-day sources of data have proven effective for dynamic volatility and tail-risk estimation and forecasting. Expected shortfall (ES) is a tail risk measure, now recommended by the Basel Committee, involving a conditional expectation that can be semi-parametrically estimated via an asymmetric sum of squares function. The conditional autoregressive expectile class of model, used to implicitly model ES, has been extended to allow the intra-day range, not just the daily return, as an input. This model class is here further extended to incorporate information on realized measures of volatility, including realized variance and realized range (RR), as well as scaled and smoothed versions of these. An asymmetric Gaussian density error formulation allows a likelihood that leads to direct estimation and one-step-ahead forecasts of quantiles and expectiles, and subsequently of ES. A Bayesian adaptive Markov chain Monte Carlo method is developed and employed for estimation and forecasting. In an empirical study forecasting daily tail risk measures in six financial market return series, over a seven-year period, models employing the RR generate the most accurate tail risk forecasts, compared to models employing other realized measures as well as to a range of well-known competitors.     

Forecasting Corporate Bankruptcy: Optimizing the Performance of the Mixed Logit Model

In recent studies, Jones and Hensher (2004 , 2005) provide an illustration of the usefulness of advanced probability modelling in the prediction of corporate bankruptcies, insolvencies and takeovers. Mixed logit (or random parameter logit) is the most general of these models and appears to have the greatest promise in terms of underlying behavioural realism, desirable econometric properties and overall predictive performance. It suggests a number of empirical considerations relevant to harnessing the maximum potential from this new model (as well as avoiding some of the more obvious pitfalls associated with its use). Using a three-state failure model, the unconditional triangular distribution for random parameters offers the best population-level predictive performance on a hold-out sample. Further, the optimal performance for a mixed logit model arises when a weighted exogenous sample maximum likelihood (WESML) technique is applied in model estimation. Finally, we suggest an approach for testing the stability of mixed logit models by re-estimating a selected model using varying numbers of Halton intelligent draws. Our results have broad application to users seeking to apply more accurate and reliable forecasting methodologies to explain and predict sources of firm financial distress better.     

House Prices and Government Spending Shocks

We show that dynamic stochastic general equilibrium (DSGE) models with housing and collateralized borrowing predict a fall in house prices following positive government spending shocks. By contrast, we show that house prices in the United States rise persistently after identified positive government spending shocks. We clarify that the incorrect house price response is due to a general property of DSGE models—approximately constant shadow value of housing—and that modifying preferences and production structure cannot help in obtaining the correct house price response. Properly accounting for the empirical evidence on government spending shocks and house prices using a DSGE model therefore remains a significant challenge.     

Methods versus substance: Measuring the effects of technology shocks

Calibration and modern (Bayesian) estimation methods for a neoclassical stochastic growth model are applied to make the case that the identification of key parameters, rather than quantitative methodologies per se, is responsible for empirical findings. For concreteness, the model is used to measure the contributions of technology shocks to the business cycle fluctuations of hours worked and output. Along the way, new insights are provided in the parameter identification associated with likelihood-based estimation, the sensitivity of likelihood-based estimation to the choice of structural shocks is assessed, and Bayesian model averaging is used to aggregate findings obtained from different DSGE model specifications.     

An empirical investigation of multiperiod tail risk forecasting models

In the context of multiperiod tail risk (i.e., VaR and ES) forecasting, we provide a new semiparametric risk model constructed based on the forward-looking return moments estimated by the stochastic volatility model with price jumps and the Cornish–Fisher expansion method, denoted by SVJCF. We apply the proposed SVJCF model to make multiperiod ahead tail risk forecasts over multiple forecast horizons for S&P 500 index, individual stocks and other representative financial instruments. The model performance of SVJCF is compared with other classical multiperiod risk forecasting models via various backtesting methods. The empirical results suggest that SVJCF is a valid alternative multiperiod tail risk measurement; in addition, the tail risk generated by the SVJCF model is more stable and thus should be favored by risk managers and regulatory authorities.     

Can VAR models capture regime shifts in asset returns? A long-horizon strategic asset allocation perspective

It is often suggested that through a judicious choice of predictors that track business cycles and market sentiment, simple vector autoregressive (VAR) models could produce optimal strategic portfolio allocations that hedge against the bull and bear dynamics typical of financial markets. However, a distinct literature exists that shows that nonlinear econometric frameworks, such as Markov switching (MS), are also natural tools to compute optimal portfolios in the presence of stochastic good and bad market states. In this paper we examine whether simple VARs can produce portfolio rules similar to those obtained under MS, by studying the effects of expanding both the order of the VAR and the number/selection of predictor variables included. In a typical stock-bond strategic asset allocation problem, we compute the out-of-sample certainty equivalent returns for a wide range of VARs and compare these measures of performance with those typical of nonlinear models for a long-horizon investor with constant relative risk aversion. We conclude that most VARs cannot produce portfolio rules, hedging demands, or (net of transaction costs) out-of-sample performances that approximate those obtained from equally simple nonlinear frameworks. We also compute the improvement in realized performance that may be achieved adopting more complex MS models and report this may be substantial in the case of regime switching ARCH.