Forecasting monthly and quarterly time series using STL decomposition

This paper is a re-examination of the benefits and limitations of decomposition and combination techniques in the area of forecasting, and also a contribution to the field, offering a new forecasting method. The new method is based on the disaggregation of time series components through the STL decomposition procedure, the extrapolation of linear combinations of the disaggregated sub-series, and the reaggregation of the extrapolations to obtain estimates for the global series. Applying the forecasting method to data from the NN3 and M1 Competition series, the results suggest that it can perform well relative to four other standard statistical techniques from the literature, namely the ARIMA, Theta, Holt-Winters’ and Holt’s Damped Trend methods. The relative advantages of the new method are then investigated further relative to a simple combination of the four statistical methods and a Classical Decomposition forecasting method. The strength of the method lies in its ability to predict long lead times with relatively high levels of accuracy, and to perform consistently well for a wide range of time series, irrespective of the characteristics, underlying structure and level of noise of the data.     

Real-time forecasting of German GDP based on a large factor model with monthly and quarterly data

This paper discusses a factor model for short-term forecasting of GDP growth using a large number of monthly and quarterly time series in real-time. To take into account the different periodicities of the data and missing observations at the end of the sample, the factors are estimated by applying an EM algorithm, combined with a principal components estimator. We discuss some in-sample properties of the estimator in a real-time environment and propose alternative methods for forecasting quarterly GDP with monthly factors. In the empirical application, we use a novel real-time dataset for the German economy. Employing a recursive forecast experiment, we evaluate the forecast accuracy of the factor model with respect to German GDP. Furthermore, we investigate the role of revisions in forecast accuracy and assess the contribution of timely monthly observations to the forecast performance. Finally, we compare the performance of the mixed-frequency model with that of a factor model, based on time-aggregated quarterly data.     

A note on interactions-driven business cycles

Within the standard Keynesian multiplier framework, extended by a micro-model of interactive formation of individual consumption propensities, we demonstrate that socioeconomic interactions can lead to cyclical fluctuations in aggregate economic activity. The underlying micro-model of direct interactions is a version of Alan Kirman’s generic opinion formation model, with an additional feedback effect from macroscopic variables on the transition probabilities. Our model engenders cyclical fluctuations of economic variables, despite the fact that neither the Keynesian multiplier model nor Kirman’s model does so on its own.     

The impact of the US on Latin American business cycles: A new approach

Business cycles in Latin America have tended to be more volatile than those in wealthier nations such as the US. Accordingly, much research has been conducted on Latin business cycles, as well as the impact of the US on such fluctuations. Some research seeks to find how “integrated” cycles are in the US and Latin America, yielding conflicting results. We apply a new method to the question of business cycle synchronization between the US and nine Latin nations. We find that in the majority of cases integration has been rising in recent years. We also find, contrary to some previous studies, that integration does not appear to be affected by either the level of trade or of capital account openness. Finally, we find that the two countries that are dollarized – Ecuador and El Salvador – appear least integrated with the US. This last finding has potentially troubling implications in terms of the ability of these nations to adjust to asymmetric shocks vis-à-vis the US.     

The use of monthly and quarterly data in an ARMA model

This paper deals with specification, prediction and length of interval between the observations in an ARMA model. An AR(1) model is found to be suitable for a specific monthly time series. From this series we construct two types of quarterly series and derive the corresponding ARMA models. The theoretical parameter values of the quarterly models, given the monthly model, are compared with the values found empirically when no monthly series exists. By using the variance of the predictor error, we assess the performance of all specifications in predicting up to one year ahead. We show that while the monthly model performs best in theory, the values computed directly from the estimates prove in our empirical example the quarterly models to be preferable in most cases where we are to predict more than one quarter ahead.     

基于新型企业观的知识流管理

本文主要讨论了在新型企业观背景下知识管理的重要意义以及如何实施等问题.     

Search for yield and business cycles

In the ultra-low interest rate environment after the financial crisis, it has been often pointed out that the “search for yield” behavior of financial institutions might have been intensifying interest rate decreases. One hypothesis to explain search for yield is that banks try to buy longer-term bonds even when they recognize negative term premiums in long-term rates because they myopically care about current portfolio income, not just expected holding-period returns. I study the potential impacts of this behavior on U.S. business cycles and long-term bond’s ex-post term premiums. I find that in an economy in which banks are exposed to the value-at-risk constraint, the existence of these myopic banks provides realistic moments of ex-post term premiums. In addition, their existence could generate higher output persistence under a productivity shock compared to an economy without them. This is because the difference between a myopic long-term bond pricing and a realized deposit rate path affects banks’ net worth. I study policy implications, too. In response to changes in the strength of banks’ capital regulation, the existence of myopic banks amplifies business cycles. Regarding monetary policy, it is necessary to consider ex-post term premiums, because ex-post term premiums tend to move in the same direction as the short-term interest rate and amplify business cycles in that direction when myopic banks exist.     

The uncertainty multiplier and business cycles

I study a business cycle model where agents learn about the fundamentals by accumulating capital. During recessions, agents invest less, and this generates noisier estimates of macroeconomic conditions and an increase in uncertainty. The endogenous increase in aggregate uncertainty further reduces economic activity and thus gives rise to a multiplier effect that amplifies aggregate fluctuations. To discipline learning dynamics, I parametrize the model so that it matches not only standard business cycle moments but also survey data on macroeconomic forecasts. I find that the uncertainty multiplier amplifies output standard deviation by 16%.     

Economic sentiment and business cycles: A spillover methodology approach

This paper is one of the latest attempts to observe and explain the relationship between the Economic Sentiment Indicator (ESI) and GDP growth. This new approach uses the Diebold and Yilmaz (2012, 2009) spillover methodology by combining the feedback relationship and the time-varying (dynamic) aspect of that same relationship. The empirical analysis is based on quarterly data over different periods in the European Union (EU) (as a whole) and in ten Central and Eastern European (CEE) countries. The main results indicate that the spillovers of shocks between ESI and GDP growth are time-varying for all observed countries, with increasing values of total spillovers in the last recession and afterwards. The direction and strength of spillover effects between economic sentiment and GDP growth are different for the analysed CEE countries. The results are robust with respect to changing the rolling window length, additional Granger causality testing and rolling correlation coefficients.     

Financial intermediation and real estate prices impact on business cycles: A Bayesian analysis

How do financial intermediation and real estate prices impinge on the business cycle? I develop a two-sector stochastic general equilibrium model with financial intermediation and real estate collateral to assess the impact of financial conditions and land prices on aggregate fluctuations. I estimate the model with Bayesian methods using a novel data set that includes U.S. macro and financial variables during the period 1975–2010. The results from the estimated model show that financial conditions have a sizable effect on the variability of investment spending, while productivity shocks are the main source of consumption fluctuations. Specifically, on the macro side, (1) financial shocks explain about three quarters of investment spending variability and one third of the variance in hours worked. On the financial side, (2) financial shocks explain most of the variability in land prices, credit spread, and aggregate net worth of the financial sector. The model also accounts for observed unconditional moments of macro and financial variables. Our quantitative results are suggestive of the impact of diverse sources of financial instability, and as such relevant for macro prudential policy analysis.     

Riskiness,endogenous productivity dispersion and business cycles

In the data, cross-sectional productivity dispersion is countercyclical at both the plant level and the firm level, see e.g. Bloom (2009). I incorporate a firm׳s choice of risk level into a model of firm dynamics with real business cycle features to explain this empirical finding both qualitatively and quantitatively. In the model, in every period, each firm chooses the investment amount and the risk level associated with a production project every period. All projects available to each firm have the same expected flow return, determined by the aggregate and idiosyncratic shocks to the firm׳s productivity, and differ from one another only in their risk. The endogenous option of exiting the market and the limited funding for new investment jointly play an important role in motivating firms׳ risk-taking behavior. The model predicts that, in each period, relatively small firms are more likely to take risk and hence exhibit a higher exit rate, and that the cross-sectional productivity dispersion, measured as the standard deviation of the realized individual component of productivity, is larger in recessions.     

Trends and cycles in economic time series: A Bayesian approach

Trends and cyclical components in economic time series are modeled in a Bayesian framework. This enables prior notions about the duration of cycles to be used, while the generalized class of stochastic cycles employed allows the possibility of relatively smooth cycles being extracted. The posterior distributions of such underlying cycles can be very informative for policy makers, particularly with regard to the size and direction of the output gap and potential turning points. From the technical point of view a contribution is made in investigating the most appropriate prior distributions for the parameters in the cyclical components and in developing Markov chain Monte Carlo methods for both univariate and multivariate models. Applications to US macroeconomic series are presented.     

Uncertainty shocks,central bank characteristics and business cycles

This paper investigates the conjecture that central bank independence and transparency moderate the negative effect of uncertainty shocks on real output. To test this conjecture, the real GDP growth rate is regressed on the interaction terms between measures of central bank characteristics and the proxy for macroeconomic uncertainty, i.e. stock market volatility. To address potential endogeneity concerns, stock market volatility is instrumented in a Two Stage Least Squares model by plausibly exogenous natural disaster, terrorist attack, political coup and revolution shocks. The estimation results provide strong evidence that central bank independence reduces the adverse effect of uncertainty shocks. There is also evidence for the moderating impact of transparency. However, due to the limited availability of transparency data, the result is less conclusive.     

Endogenous business cycles caused by nonconvex costs and interactions

This paper shows that endogenous business cycles (inventory cycles) arise from a combination of nonconvex costs and economic interactions among firms. At the micro level, firm behavior is characterized by lumpiness, and the standard production-smoothing theory is empirically rejected. To account for this, a nonconvex cost function is assumed in our model. It might be expected that even if the microeconomic behavior is lumpy, the effect disappears at the aggregate level because of the law of large numbers. However, we show that if there exist interactions among firms, a regular endogenous cycle emerges at the aggregate level given that the degree of the interaction effect exceeds a critical point. That is, the randomly behaving microeconomic agents generate deterministic collective behavior via interactions. This offers an explanation for the Kitchin cycle.     

Risk aversion and business cycles: An empirical analysis

Unlike investors, who tend to maintain highly-diversified portfolios, private entrepreneurs usually lack access to complete risk-pooling for idiosyncratic risks, thus more directly internalize the cost of volatility. Risk aversion, however, modifies the optimal contract between entrepreneurs and lenders by incorporating the risk premium that entrepreneurs demand for the uninsurable risk: the private equity premium. Consequently, real shocks tend to be amplified as changes in entrepreneurs’ net worth affect the private equity premium and so the rental rate of capital, investment and output. This theoretical framework suggests that economies where the private entrepreneurial sector is a relatively larger, and therefore more vulnerable to uninsurable risk, all else equal, should present higher volatility. I test this prediction by (1) conducting a simple reduced-form analysis that shows that output volatility is negatively associated with the relative importance of the corporate vs. the privately-held sector; and (2) estimating the model's structural parameters. Intuitively, countries where private entrepreneurs are predominant and so risk aversion is likely to impose stronger impacts, positive risk aversion coefficients should be found. Results suggest that risk aversion is empirically more relevant for economies like Argentina, Brazil, Chile, Korea, Mexico and Thailand than for Canada, France, Germany, the U.K. and the U.S.     

Fiscal policies and business cycles in an enlarged euro area

This paper compares the cyclical properties of fiscal policies across the 12 original eurozone countries and the future members from Central and Eastern Europe. For the sample period 1995–2005, the fiscal balance exhibits less inertia and is more counter-cyclical in Central and Eastern European countries than in members of the eurozone. The main differences arise from the revenue side. Differences in the formation of fiscal policy between current and future eurozone countries decrease over time. Both autonomous and counter-cyclical fiscal policies have little or no effect on cyclical variability in the eurozone countries, while such policies appear to be effective in Central and Eastern European countries.     

The yearend effect in stock returns over business cycles: A technical note

This paper provides evidence in support of the claim that the well-knownJanuary effect is influenced by the stage of the business cycle. Using monthly data for the S&P Composite Index for the period from November 1948 through December 1988 and the standard methodology for seasonal anomalies, the authors show that theJanuary effect is present during the entire period examined as well as in the expansionary phases of that period. However, its existence was not detected during the contractionary phases of that period.     

Forecasting an accumulated series based on partial accumulation II: A new Bayesian method for short series with stable seasonal patterns

The problem of forecasting a time series with only a small amount of data is addressed within a Bayesian framework. The quantity to be predicted is the accumulated value of a positive and continuous variable for which partially accumulated data are available. These conditions appear in a natural way in many situations. A simple model is proposed to describe the relationship between the partial and total values of the variable to be forecasted assuming stable seasonality, which is specified in stochastic terms. Analytical results are obtained for both the point forecast and the entire posterior predictive distribution. The proposed technique does not involve approximations. It allows the use of non-informative priors so that implementation may be automatic. The procedure works well when standard methods cannot be applied due to the reduced number of observations. It also improves on previous results published by the authors. Some real examples are included.