Analysis of non-stationary dynamics in the financial system

Novel data-driven analyses, appropriate for detecting economic instability in non-stationary time series, are developed using functional principal component analysis (fPCA) and Synchrosqueezing. fPCA is applied in a new way, aggregating multiple financial time series to identify periods of macroeconomic instability. Synchrosqueezing, a technique which generates a time-series’ time-dependent spectral decomposition, is modified to develop a new quantitative measure of local dynamical changes and structural breaks. The merit of this integrated technique is demonstrated by analyzing financial data from 1986 to 2012 that includes equity indices, securities and commodities, and foreign exchange. Both procedures successfully detect key historic periods of instability. Moreover, the results reveal distinctions between periods of long-term gradual change in addition to structural breaks. These tools offer new insights into the analysis of financial instability.     

A spectral decomposition for structural VAR models

Based on structural VARs, this paper proposes a spectral decomposition which allows to infer the effects of changes in one variable on the other variables in the frequency domain. It is shown that there is a close relationship between this concept and conventional forecast error variance decomposition techniques for VARs. An empirical example demonstrates the usefulness of this additional tool in analyzing the relationships among time series.     

Testing for individual and time effects in the two-way error component model with time-invariant regressors

In this paper, we focus on testing for individual and time effects in the two-way error component model with time-invariant regressors. We present the so-called FEF estimators when time-invariant regressors are exogenous and the FEF-IV estimators when one or more of time-invariant variables are endogenous, and obtain their asymptotic properties under some mild conditions. In the light of the moment-based test methods of Wu and Li (2014), we construct several tests for the existence of individual and time effects in the two-way error component model with time-invariant regressors. The resulting tests can be shown to have some desired properties as follows: they do not need any distributional assumptions on the error components; they do not require any assumptions on the correlation among the two random effects and the time-varying regressors; they are robust to the presence of one effect when the other one is tested. Simulation study and real data analysis are carried out for illustration of the above.     

Changes in the relationship between short‐term interest rate,inflation and growth: evidence from the UK, 1820–2014

This paper examines the dynamic relationship between interest rates, inflation and economic growth using a long dataset for the UK. The approach adopted enables us to identify structural breaks in the dynamic system (vector autoregression (VAR)). We find interest rates respond much more strongly to growth and inflation over recent decades, and forecast error variance decomposition analysis indicates there is increasing interconnectedness between the variables in recent years. Economic policymakers need to carefully monitor the linkages between these variables and be prepared to adjust their monetary policy tools when faced with structural changes.     

Error-correction based panel estimates of the demand for money of selected Asian countries with the extreme bounds analysis

This paper uses the extreme bounds analysis (EBA) of Leamer (1983, 1985) to analyze the robust determinants of the demand for money in a panel of 17 Asian countries for the period 1970 to 2009. These robust determinants are found to be unit root variables. Therefore, cointegration between these variables is tested with a recent time series panel method developed by Westerlund (2007). This method uses the error-correction formulation and has more power against the null of no cointegration. The results show that there is a well-defined long-run demand for money. Using the lagged error correction term from the estimated cointegrating equation, the short-run dynamic relationships are estimated. This paper, thus, suggests some useful guidelines to estimate other relationships with panel data.     

What can we learn from revisions to the Greenbook forecasts?

Although there have been many evaluations of the Federal Reserve’s Greenbook forecasts, we analyze them in a different dimension. We examine the revisions of these forecasts in the context of fixed event predictions to determine how new information is incorporated in the forecasting process. This analysis permits us to determine if there was an inefficient use of information in the sense that the forecast revision has predictive power for the forecast error. Research on forecast smoothing suggests that we might find a positive relationship between the forecast error and the forecast revision. Although we do find for some variables and horizons the Fed’s forecast errors are predictable from its forecast revisions, there is no evidence of forecast smoothing. Instead the revisions sometimes have a negative relationship with the forecast error, suggesting in these cases that the Fed may be over-responsive to new information.     

Modeling nonlinear Granger causality between the oil price and U.S. dollar: A wavelet based approach

In this paper, we use a wavelet approach to study the linear and nonlinear Granger causality between the real oil price and the real effective U.S. Dollar exchange rate. Instead of analyzing the time series at their original level, as it is usually done, we first decompose the two macroeconomic variables at various scales of resolution using wavelet decomposition and then we study the relationships among the decomposed series on a scale by scale basis. A major finding of this paper is that the linear and nonlinear causal relationships between the real oil price and the real effective U.S. Dollar exchange rate vary over frequency bands as it depends on the time scales. Indeed, there is a strong bidirectional causal relationship between the real oil price and the real dollar exchange rate for large time horizons, i.e. corresponding to fundamentalist traders, especially fund managers and institutional investors. But, for the first frequency band which corresponds to a class of traders whom investment horizon is about 3-months and whom trading is principally speculative (noise traders), the causality runs only from the real oil prices to real effective U.S dollar exchange rate.     

Official fiscal forecasts in EU member states under the European Semester and Fiscal Compact – An empirical assessment

The efficacy of official forecasts in the EU has been under the spotlight since the introduction of the euro, with biases widely reported prior to the 2008–12 financial and sovereign bond market crisis. Changes to the EU fiscal rules and procedures, in the form of the European Semester and Fiscal Compact, in the early 2010s were adopted to improve forecasting, including through providing a role for independent fiscal institutions. Using data for 22 countries between 2013 and 2019, this paper shows that, despite these changes, biases, of a pessimistic form, remain in forecasts of budget balance and output variables in Stability and Convergence Programmes and the European Commission's Spring Forecasts. Econometric analysis indicates forecast errors in both the headline budget balance and the structural budget balance being explained by forecast errors in output variables and by EU fiscal rule requirements. Member states under an excessive deficit procedure provide optimistic headline budget balance forecasts compared to non-EDP countries, while those that have not met their medium-term objective report smaller forecast errors for the structural budget balance. Independent fiscal institutions are linked to a smaller bias to forecasts of the structural budget balance but have no effect on the forecast errors of the headline budget balance.     

Rationality and the Risk Premium on the Australian dollar

A Model of the forward rate error of the USD/AUD spot exchange rate is fitted to daily data for the period 15th December 1983 to 31st December 1991. This provides a data set of 2034 daily trading observations. Explanations of the forecast error include a risk premium represented by a constant plus the conditional variance generated from a GARCH (1,1)-M analysis of the error process and information variables in the form of lagged forward rate errors. The following conclusions are drawn form estimates for the full sample: the USD/AUD spot rate is subject to a constant premium: there is little evidence to support a time varying component and the market is influenced by lagged forward errors. Sub period estimation confirms these results, although a time varying premium is evident prior to the February 1985 depreciation. The economic implications of these findings are discussed. [F31]     

Measuring business cycles: A wavelet analysis of economic time series

Multiresolution wavelet analysis is a natural way to decompose an economic time series into trend, cycle, and noise. The method is illustrated with GDP data. The business-cycle component of the wavelet-filtered series closely resembles the series filtered by the approximate bandpass filter.     

Estimates of real economic activity in Switzerland, 1886–1930

This paper uses annual data spanning 1870 to 1930 on a set of variables correlated with business conditions to construct an index of real economic activity in Switzerland. We extract an estimate of the common component of the data series using principal components analysis and the unobservable variables approach proposed by Stock and Watson (1989, 1991). The resulting index is similar to that constructed by Andrist et al. (2000) but displays more variation over time and is available for a longer time period. Moreover, it is less volatile and covers a longer time period in the 20th century than the estimate by Ritzmann-Blickenstorfer (1996).First version received: May 2002/Final version received: September 2004The views expressed in this paper are solely our own and are not necessarily shared by the institutions we are affiliated with. We thank the Editor and the anonymous referees for helpful comments (in particular for finding serious errors in the principal components analysis), and Jim Stock and Mark Watson for detailed explanations of their estimation strategy.     

For they know not what they do: an analysis of monetary policy during the Great Moderation

In this article, we develop an empirical framework to show the importance of money during the Great Moderation, while accounting for the fact that monetary policy was exclusively conducted through interest rates. We estimate the impulse response functions and forecast error variance decomposition derived from a structural VAR with a least absolute shrinkage and selection operator–based lag selection. The variance decomposition suggests that a substantial component of macroeconomic variation has been driven by shocks to the money market, which were not only unintended by the Federal Reserve, but worse passed unnoticed allowing those shocks to accumulate over time.     

Usefulness of linear transformations in multivariate time-series analysis

Much progress has been made in recent years in multivariate time-series analysis. In this paper we summarize some of the methodological developments that are particularly relevant to empirical economics and highlight especially the usefulness of linear transformations in analyzing multivariate time series. The topics considered include vector ARMA models, principal component analysis, scalar component models, canonical correlation analyses, co-integration, and unit-root tests. We illustrate the methods considered by an example using Taiwan's interest-rate series and provide critiques of these developments.     

中国经济周期的非对称性和相关性研究

本文利用时间序列模型等计量方法 ,对一些主要宏观经济变量序列进行随机分解和相关性分析 ,对经济周期的非对称性和长尾性质进行大量实证检验 ,对一些主要宏观经济变量序列之间的扰动和关联进行了计量分析 ,从中不仅识别了经济波动当中的各种非对称类型 ,而且分析了产生非对称的原因。本文分析得到的一些重要检验结果 ,可以作为描述中国经济波动的重要典型化事实 ,可以用于进一步分析和判断中国经济的运行趋势 ,检验和校正相应的经济理论。本文认为中国经济周期的非对称 ,主要是由固定资产投资、财政政策和货币政策的非对称性造成的 ,而价格水平和总需求等因素却保持了比较明显的稳定性。通过经济周期的非对称分析和经济变量周期成分之间的关联性分析 ,我们也具体地分析了宏观经济政策的有效性、方向性和时滞性     

Revisiting the inflation–output gap relationship for France using a wavelet transform approach

The purpose of the paper is to revisit the inflation–output gap relationship using a new approach known as the wavelet transform. This approach combines the classical time series analysis with frequency domain analysis and presents the advantages of assessing the co-movement of the two series in the context of both time and frequencies. Using discrete and continuous wavelet methodologies for the study of the inflation–output gap nexus in the case of France, we determine that the output gap is able to predict the inflation dynamics in the short- and medium-runs, and these results have important implications to the Phillips curve theory. More precisely, we discovered that in a discrete wavelet framework, the short- and medium-term fluctuations of both variables are more closely correlated, whereas the continuous wavelet analysis states that the output gap leads inflation in short- and medium-runs.     

Time scale evaluation of economic forecasts

A maximal overlap discrete wavelet transform is used to obtain time scale decompositions of economic forecasts and their errors. The generated time scale components can be used in loss measures and tests for comparing forecast accuracy to evaluate whether the forecasts accurately capture the cyclical features of the data.     

Does Ecological Footprint Impede Economic Growth? An Empirical Analysis Based on the Environmental Kuznets Curve Hypothesis

This paper examines the relationship between income and environmental quality using environmental Kuznets curve (EKC) hypothesis. The hypothesised link is tested using time‐series analysis of 22 countries over the period 1961–2011. The degree of environmental impacts of economic activity is measured using ecological footprint (EF) per capita as explanatory variable, while real gross domestic product (GDP) per capita and its quadratic and cubic forms are used as predictor variables in these countries. First, the EKC hypothesis is tested through examining the relationship between EF and GDP using linear, quadratic and cubic functions. Further, the long‐run relationship between EF and GDP is investigated using a vector error correction model. It was found that there is a cointegrated relationship between the variables in almost all countries, which was statistically significant, and EKC supported in 10 countries. Additionally, almost all error correction terms are correct in sign and are significant, which implies that some percentage of disequilibria in EF in the previous year adjusts back to the long‐run equilibrium in the current year. Therefore, an efficient trade‐off between environmental protection and economic benefits should be taken, and EF should be reduced through changing consumption patterns, improving the efficiency of use of resources and cleaner technology choices.     

The determinants of FDI in Turkey: A Markov Regime-Switching approach

This paper considers movements of Foreign Direct Investments (FDI) in Turkey, and therefore, to understand the dynamics of FDI, runs several nonlinear FDI equations in which the basic determinants of FDI in Turkey are determined through Markov Regime-Switching Models (MSMs). The statistical properties of Markov Regime-Switching time series models are more desirable than those of conventional time series or panel regression models. Through these properties of MSMs, i) one can observe structural changes, if they exist, in FDI equations through time, ii) if, in fact, the true FDI regression equation follows a nonlinear relationship, MSMs fit data better than the linear models. This paper eventually follows maximum likelihood methodology of Markov Regime-Switching Model (MSM) to search for the possible structural changes in level and/or trends and possible changes in parameters of independent variables of FDI–MSM equations through the transition probabilities. In conclusion, this paper yields the outcome that Turkish FDI growth equation has significant structural changes in level and trend and that has significant coefficient shifts in explanatory variables. These explanatory variables are Turkish GDP Growth, Labor Cost, the Electricity Price Growth, the growth in average prices of High Sulphur Fuel Oil, Cooking Coal, Steam Coal and Natural Gas, Export Growth, Import Growth, Discount Rate and Country Risk Indexes for Turkey, US and EU, respectively, within the time interval from 1988 first quarter to 2010 second quarter.     

Using the Haar wavelet transform in the semiparametric specification of time series

Using theoretical arguments for nonparametric wavelet estimation, we devise regression-based semiparametric wavelet estimators to dissect linear from nonlinear effects in a time series. The wavelet estimators localize in both time and frequency so that distortion due to outliers is lessened. Our regression-based approach also lends itself to ease of replication, clarity, flexibility, timeliness and statistical validity. We demonstrate the efficacy of the approach via rolling regressions on time series of quarterly U.S. GDP growth rates, monthly Hong Kong/ U.S. exchange rates, weekly 1-month commercial interest rates and daily returns on the S&P 500.     

Growth accounting for the euro area

This article is concerned with the estimation of euro area potential output growth and its decomposition according to the sources of growth. The growth accounting exercise is based on a multivariate structural time series model which combines the decomposition of total output according to the production function approach with price and wage equations that embody Phillips-type relationships linking inflation and nominal wage dynamics to the output gap and cyclical unemployment, respectively. Assuming a Cobb?CDouglas technology with constant returns to scale, potential output results from the combination of the trend levels of total factor productivity (TFP) and factor inputs, capital and labour (hours worked), which is decomposed into labour intensity (average hours worked), the employment rate, the participation rate and population of working age. The nominal variables (prices and wages) play an essential role in defining the trend levels of the components of potential output, as the latter should pose no inflationary pressures on prices and wages. The structural model is further extended to allow for the estimation of potential output growth and the decomposition according to the sources of growth at different horizons (long run, medium run and short run); in particular, we propose and evaluate a model-based approach to the extraction of the low-pass component of potential output growth at different cut-off frequencies. The approach has two important advantages: the signal extraction filters have an automatic adaptation property at the boundaries of the sample period, so that the real-time estimates do not suffer from what is often referred to as the ??end-of-sample bias??. Secondly, it is possible to assess the uncertainty of potential output growth estimates with different degrees of smoothness.