Forecasting the exchange rate using nonlinear Taylor rule based models

This research utilises a non-linear Smooth Transition Regression (STR) approach to modelling and forecasting the exchange rate, based on the Taylor rule model of exchange rate determination. The separate literatures on exchange rate models and the Taylor rule have already shown that the non-linear specification can outperform the equivalent linear one. In addition the Taylor rule based exchange rate model used here has been augmented with a wealth effect to reflect the increasing importance of the asset markets in monetary policy. Using STR models, the results offer evidence of non-linearity in the variables used and that the interest rate differential is the most appropriate transition variable. We conduct the conventional out-of-sample forecasting performance test, which indicates that the non-linear models outperform their linear equivalents as well as the non-linear UIP model and random walk.     

Smooth-transition error-correction in exchange rates

Recent research has increasingly suggested that exchange rates may be characterized by non-linear behavior. This paper examines whether such non-linear behavior is evident, not in rates themselves, but in the adjustment of rates back to fundamental equilibrium. Thus, we examine whether a series of four spot and forward exchange rates exhibit smooth transition non-linear error-correction dynamic behavior. Our results are supportive of this model, particularly in-sample, and suggest some salient differences in the mean-reverting behavior of spot and forward rates, which may be of use to policy authorities and model builders. However, out-of-sample forecast errors between the two models appear insignificantly different from each other.     

Multilateral adjustment,regime switching and real exchange rate dynamics

The purpose of this paper is to examine the role of multilateral adjustment to U.S. external imbalances in driving bilateral real exchange rate movements by developing a new regime-switching model that consists of a Markov-switching model with a time-varying transition matrix that depends on a threshold variable. Consequently, the dynamics of the real exchange rate can be modeled in the context of two regimes: one in which multilateral adjustment to large U.S. external imbalances is an important factor driving movements in the real exchange rate and the second in which the real exchange rate is driven mainly by country-specific macroeconomic fundamentals. We apply this model to the bilateral real Canada–U.S. dollar exchange rate and compare its performance to several other alternative models. All of the models are estimated using a Bayesian approach. Our findings suggest that during periods of large U.S. imbalances, an exchange rate model for the real Canada–U.S. dollar exchange rate should allow for multilateral adjustment effects.     

Asymmetric effects of exchange rate changes on the Malaysia-China commodity trade

Previous research that considered the response of the trade balance between Malaysia and China to exchange rate changes used a linear model and did not find any significant long-run link. Suspecting that the results suffer from aggregation bias as well as ignoring nonlinear adjustment of the exchange rate, we consider the trade balance of 59 industries that trade between the two countries and use a nonlinear ARDL model to show that almost a third of the industries are affected by ringgit depreciation against the yuan, in an asymmetric manner. The largest industry, which accounts for more than 25% of the trade, is found to benefit from ringgit depreciation while not being hurt by appreciation. In total, 15 industries that account for 40% of the trade enjoy this property.     

A non-linear model of the real US/UK exchange rate

This paper provides a framework for building and estimating non-linear real exchange rate models. The approach derives the stationary distribution from a continuous time error correction model and estimates this by MLE methods. The derived distribution exhibits a wide variety of distributional shapes including multimodality. The main result is that swings in the US/UK rate over the period 1973:3 to 1990:5 can be attributed to the distribution becoming bimodal with the rate switching between equilibria. By capturing these changes in the distribution, the non-linear model yields improvements over the random walk, the speculative efficiency model, and Hamilton's stochastic segmented trends model.     

Import demand behavior in Africa: Some new evidence

Some African economies have experienced increases in the level of their foreign exchange reserves as well as increases in their import volume. Theory suggests that as the level of exchange reserves increases, it may affect the demand for imports since more funds will be available for imports. This paper examines import demand behavior in three African economies, namely Kenya, Nigeria and South Africa. An empirical analysis of import demand behavior is presented, based on the dynamic error-correction model, which allows an explicit parameterized division of effects into long-run influences, short-term adjustment and error-correction term. It uses econometric techniques organized around Johansen and Harris–Inder cointegration analyses; fully modified OLS, dynamic OLS and non-linear OLS to estimate long-run import demand functions.     

Further Evidence on PPP Adjustment Speeds: the Case of Effective Real Exchange Rates and the EMS*

Two different approaches intend to resolve the ‘puzzling’ slow convergence to purchasing power parity (PPP) reported in the literature [see Rogoff (1996) , Journal of Economic Literature, Vol. 34.] On the one hand, there are models that consider a non‐linear adjustment of real exchange rate to PPP induced by transaction costs. Such costs imply the presence of a certain transaction band where adjustment is too costly to be undertaken. On the other hand, there are models that relax the ‘classical’ PPP assumption of constant equilibrium real exchange rates. A prominent theory put together by Balassa (1964, Journal of Political Economy, Vol. 72) and Samuelson (1964 Review of Economics and Statistics, Vol. 46) , the BS effect, suggests that a non‐constant real exchange rate equilibrium is induced by different productivity growth rates between countries. This paper reconciles those two approaches by considering an exponential smooth transition‐in‐deviation non‐linear adjustment mechanism towards non‐constant equilibrium real exchange rates within the EMS (European Monetary System) and effective rates. The equilibrium is proxied, in a theoretically appealing manner, using deterministic trends and the relative price of non‐tradables to proxy for BS effects. The empirical results provide further support for the hypothesis that real exchange rates are well described by symmetric, nonlinear processes. Furthermore, the half‐life of shocks in such models is found to be dramatically shorter than that obtained in linear models.     

A univariate analysis of EMS exchange rates using a target zone model

The models in the literature on exchange-rate target zones imply a non-linear time series model for the exchange rate. We show how the parameters of such models can be estimated and develop Maximum Likelihood and Method of Simulated Moments estimators for the target zone model of Krugman (1991). The Maximum Likelihood estimator is based on a computationally attractive approximation to the exact predictive density of the continuous time model. Monte Carlo experiments are used to assess the properties of this estimator. In the empirical part we estimate the model with data on recent EMS exchange rates. We find that the Krugman (1991) target zone model is not able to explain the full observed kurtosis and conditional heteroscedasticity of the exchange-rate returns.     

Distinguishing between stochastic and deterministic behavior in high frequency foreign exchange rate returns: Can non-linear dynamics help forecasting?

This paper investigates the dynamic properties of high frequency foreign exchange rate returns. Using hourly data for four exchanges rates, the British Pound, the Deutschemark, the Japanese Yen and Swiss Franc, we attempt to differentiate between stochastic and deterministic behavior in hourly rates of returns. While the autocorrelation coefficients and the Brock-Dechert-Scheinkman test point to the presence of some non-linear dependence, correlation dimension estimates reveal little evidence in favor of low-dimensional chaos. The analysis appears to support the view that although it is not possible to exploit deterministic non-linear dependence in exchange rate time series in order to improve short-term forecasting, non-linear stochastic models can be used for conditional volatility forecasts.     

Nonlinear adjustment,purchasing power parity and the role of nominal exchange rates and prices

Although the literature on purchasing power parity (PPP) is rich in controversy, the relative contribution of prices and nominal exchange rates to real exchange rate movements which restore PPP disequilibria has rarely been put under any close scrutiny. This paper as a first step applies a cointegrated VAR framework to test for stationary real exchange rates and linear adjustments in prices and nominal exchange rates. As a second step, ESTR error correction models are fitted to test whether nonlinear error correctional behaviour characterizes the data. The results clearly indicate that the nominal exchange rate is responsible for the nonlinear mean reverting behaviour in real exchange rates and also mainly drives overall adjustment. Applying dynamic stochastic simulations based on the estimated models, this study also confirms recent results that the half-life times of real exchange rate shocks are significantly smaller than the consensus benchmark of 3–5 years.     

Understanding Price Stickiness: Firm‐level Evidence on Price Adjustment Lags and Their Asymmetries

We study the speed of price reactions to positive and negative demand and cost shocks. Our findings suggest that price adjustment lags vary in line with the predictions of optimal price setting models. Moreover, we find that the firms' reactions are asymmetric, and that these asymmetries cannot be fully explained by any single theoretical model of asymmetric price adjustment. Overall, these results suggest that the reaction to monetary policy shocks may depend on which firms or sectors are particularly affected by them and, therefore, that richer models are needed to fully understand the effects of monetary policy.     

Non-linear exchange rate relationships: An automated model selection approach with indicator saturation

This paper examines whether the explanatory power of exchange rate models can be improved by allowing for cross-country asymmetries and non-linear effects of fundamentals. Both appear to be crucial. The samples include the USD versus pound and yen from 1982:10 to 2013:10, and automated model selection is conducted with indicator saturation. Several non-linear effects are significant at 1%. Further, many of the indicators present in the linear models are eliminated once allowing for non-linearities; suggesting some of the structural breaks found in previous work were an artifact of the misspecified linear functional form. These conclusions are robust to estimation using principal components.     

U.S.-German commodity trade and the J-curve: New evidence from asymmetry analysis

Since the introduction of the nonlinear ARDL approach and asymmetric cointegration and error-correction modeling, old relations are receiving renewed attention, and the link between the trade balance and the exchange rate is no exception. We add to this new literature by using industry-level data from 59 2-digit industries that trade between the U.S. and Germany. We find that when the old approach of the linear model was used, the real dollar-euro rate had short-run effects in 17 industries that lasted into the long run in 26 industries. However, when the nonlinear model was estimated, we found short-run asymmetric effects of exchange rate changes in 49 industries, which lasted into long-run asymmetric effects in 28 industries. The J-curve effect was supported in a total of 18 industries.     

Time-varying risk aversion and renminbi exchange rate volatility: Evidence from CARR-MIDAS model

In this paper, we investigate the relation between time-varying risk aversion and renminbi exchange rate volatility using the conditional autoregressive range-mixed-data sampling (CARR-MIDAS) model. The CARR-MIDAS model is a range-based volatility model, which exploits intraday information regarding the intraday trajectory of the price. Moreover, the model features a MIDAS structure allowing for time-varying risk aversion to drive the long-run volatility dynamics. Our empirical results show that time-varying risk aversion has a significantly negative effect on the long-run volatility of renminbi exchange rate. Moreover, we observe that both intraday ranges and time-varying risk aversion contain important information for forecasting renminbi exchange rate volatility. The range-based CARR-MIDAS model incorporating time-varying risk aversion provides more accurate out-of-sample forecasts of renminbi exchange rate volatility compared to a variety of competing models, including the return-based GARCH, GARCH-MIDAS and GARCH-MIDAS incorporating time-varying risk aversion as well as range-based CARR, CARR-MIDAS and heterogeneous autoregressive (HAR), for forecast horizons of 1 day up to 3 months. This result is robust to alternative risk aversion measure, alternative MIDAS lags as well as alternative out-of-sample periods. Overall, our findings highlight the value of incorporating intraday information and time-varying risk aversion for forecasting the renminbi exchange rate volatility.     

Internal and external transmissions of monetary and fiscal policies in the EMU

With the introduction of Economic and Monetary Union (EMU), the sovereignty of national monetary institutions has been replaced by a common monetary institution, the European Central Bank (ECB) and national currencies have been replaced by a common currency, the euro. EMU therefore implies the loss of national monetary policy autonomy and internal exchange rate flexibility inside the EMU area. However, external exchange rate adjustment, i.e. adjustment of the euro exchange rate, remains a feasible adjustment mechanism. This paper analyses how internal and external exchange rate flexibility affect macroeconomic adjustment in EMU and non-EMU countries. To do so, a model is constructed in which three countries interact: two countries that decide to form a monetary union and a third country that does not participate in the monetary union. Numerical simulations of a representative example are used to characterise the adjustment dynamics induced by monetary and fiscal policies before and after the start of the EMU.     

Pricing‐to‐Market and Market Structure*

Stimulated by imperfect competition/sticky prices framework of the new open economy macroeconomics, empirical research has reconsidered the role of exchange rates in international adjustment. This paper reassesses the link between exchange rates and traded good prices by estimating pricing‐to‐market equations for the five main euro area countries over the period 1990–99. We minimize selection biases by keeping all manufacturing products and all destination markets and show that exchange rate pass‐through (ERPT) is much larger, almost complete, than previously estimated. Thanks to a huge variability in terms of exchange rate variations, products and destination markets, we can map differences in ERPT into market structures and, at the same time, reconcile our results with the empirical literature. We find that ERPT is highly incomplete for sales by oligopolistic industries into advanced economies, indeed in the order of 50–60% as previously estimated. ERPT is instead almost complete in emerging and developing economies where therefore exchange rate movements can help adjust external imbalances. We also find that ERPT is largely asymmetric: it is almost complete after an appreciation of the exporter's currency, rather incomplete after a depreciation. This result is very robust across specifications.     

The impact of the shutdown policy on the asymmetric interdependence structure and risk transmission of cryptocurrency and China’s financial market

By taking Bitcoin, Litecoin, and China’s gold and RMB/US dollar exchange rate market as research objects, this paper apply the MF-ADCCA and time-delayed DCCA methods to study the impact of China’s mainland shutdown of cryptocurrencies trading on the non-linear interdependent structure and risk transmission of cryptocurrencies and its financial market. Empirical results show that the cross-correlation between cryptocurrencies and China’s financial market has a long memory and asymmetric multifractal characteristics. After the shutdown, the long memory between cryptocurrencies and Chinese gold has weakened, and the long memory between cryptocurrencies and the RMB/US dollar exchange rate market was strengthened. China’s shutdown policy has a certain risk prevention effect. Specifically, after the implementation of the policy, the risk transmission of cryptocurrencies to China’s financial market has weakened, but the influence of China’s financial market has gradually strengthened.     

Assessing DSGE model nonlinearities

We develop a new class of time series models to identify nonlinearities in the data and to evaluate DSGE models. U.S. output growth and the federal funds rate display nonlinear conditional mean dynamics, while inflation and nominal wage growth feature conditional heteroskedasticity. We estimate a DSGE model with asymmetric wage and price adjustment costs and use predictive checks to assess its ability to account for these nonlinearities. While it is able to match the nonlinear inflation and wage dynamics, thanks to the estimated downward wage and price rigidities, these do not spill over to output growth or the interest rate.     

The conditional heteroscedasticity of the yen–dollar exchange rate

This paper examines the conditional heteroscedasticity of the yen–dollar exchange rate. A model is constructed by extending the asymmetric power autoregressive conditional heteroscedasticity model to a process that is fractionally integrated. It is found that, unlike the equity markets, the appreciation and depreciation shocks of the yen against the dollar have similar effects on future volatilities. Although the results reject both the stable and the integrated models, our analysis of the response coefficients of the past shocks and the application of the models to the estimation of the capital requirements for trading the currencies show that there are no substantial differences between the fractionally integrated models and the stable models. © 1998 John Wiley & Sons, Ltd.