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.     

Forecast content and content horizons for some important macroeconomic time series

Abstract.  The information content of statistical forecasts of approximately stationary quantities tends to decline as the forecast horizon increases, and there exists a maximum horizon beyond which forecasts cannot provide discernibly more information about the variable than is present in the unconditional mean (the content horizon ). The pattern of decay of forecast content (or skill) with increasing horizon is well known for many types of meteorological forecasts; by contrast, little generally accepted information about these patterns or content horizons is available for economic variables. In this paper we estimate content horizons for a variety of macroeconomic quantities; more generally, we characterize the pattern of decay of forecast content as we project farther into the future. We find a wide variety of results for the different macroeconomic quantities, with models for some quantities providing useful content several years into the future, for other quantities providing negligible content beyond one or two months or quarters.     

Multi-horizon inflation forecasts using disaggregated data

In this paper we use multi-horizon evaluation techniques to produce monthly inflation forecasts for up to twelve months ahead. The forecasts are based on individual seasonal time series models that consider both, deterministic and stochastic seasonality, and on disaggregated Consumer Price Index (CPI) data. After selecting the best forecasting model for each index, we compare the individual forecasts to forecasts produced using two methods that aggregate hierarchical time series, the bottom-up method and an optimal combination approach. Applying these techniques to 16 indices of the Mexican CPI, we find that the best forecasts for headline inflation are able to compete with those taken from surveys of experts.     

German business cycle forecasts, asymmetric loss and financial variables

We examine the efficiency of German forecasts for output growth and inflation allowing for an asymmetric loss function of the forecasters. We find the loss of output growth forecasts to be approximately symmetric while there is an asymmetry in the loss of the inflation forecasts. The information of financial variables seems to be adequately incorporated into the output forecasts but to a lesser extent into the inflation forecasts.     

Forecasting the probability of US recessions: a Probit and dynamic factor modelling approach

Abstract Quantifying the probability of U.S. recessions has become increasingly important since August 2007. In a data‐rich environment, this paper is the first to apply a Probit model to common factors extracted from a large set of explanatory variables to model and forecast recession probability. The results show the advantages of the proposed approach over many existing models. Simulated real‐time analysis captures all recessions since 1980. The proposed model also detects a significant jump in the next six‐month recession probability based on data up to November 2007, one year before the formal declaration of the recent recession by the NBER.     

Evaluating the effect of monetary policy on unemployment with alternative inflation forecasts

This paper explores the role that inflation forecasts play in the uncertainty surrounding the estimated effects of alternative monetary rules on unemployment dynamics in the euro area and the US. We use the inflation forecasts of 8 competing models in a standard Bayesian VAR to analyse the size and the timing of these effects, as well as to quantify the uncertainty relative to the different inflation models under two rules. The results suggest that model uncertainty can be a serious issue and strengthen the case for a policy strategy that takes into account several sources of information. We find that combining inflation forecasts from many models not only yields more accurate forecasts than those of any specific model, but also reduces the uncertainty associated with the real effects of policy decisions. These results are in line with the model-combination approach that central banks already follow when conceiving their strategy.     

Inflation targeting in Latin America: Empirical analysis using GARCH models

This paper studies inflation dynamics in eight Latin American countries, some of which have adopted formal inflation targets (IT) as their monetary policy frameworks. We analyze the possible benefits associated with IT, not only in terms of inflation level and volatility, but also regarding other nonlinear characteristics of these series, such as volatility persistence or the fulfillment of the Friedman hypothesis. To describe inflation dynamics we use an unobserved components model, where each component can follow a GARCH type process. Once we estimate the model, the main findings of the empirical exercise confirm the favorable performance of IT.     

Nonlinear interest rate reaction functions for the UK

We empirically analyze Taylor-type equations for short-term interest rates in the United Kingdom using quarterly data from 1970Q1 to 2006Q2. Starting from strong evidence against a simple linear Taylor rule, we model nonlinearities using logistic smooth transition regression (LSTR) models. The LSTR models with time-varying parameters consistently track actual interest rate movements better than a linear model with constant parameters. Our preferred LSTR model uses lagged interest rates as a transition variable and suggests that in times of recessions the Bank of England puts more weight on the output gap and less so on inflation. A reverse pattern is observed in non-recession periods. Parameters of the model change less frequently after 1992, when an inflation target range was announced. We conclude that for the analysis of historical monetary policy, the LSTR approach is a viable alternative to linear reaction functions.     

Real-time data revisions and the PCE measure of inflation

This paper tracks data revisions in the Personal Consumption Expenditure using the exclusions-from-core inflation persistence model. Keeping the number of observations the same, the regression parameters of earlier vintages of real-time data, beginning with vintage 1996:Q1, are tested for coincidence against the regression parameters of the last vintage of real-time data, used in this paper, which is vintage 2008:Q2 in a parametric and two nonparametric frameworks. The effects of data revisions are not detectable in the vast majority of cases in the parametric model, but the flexibility of the two nonparametric models is able to utilize the data revisions.     

Assessing the impact of the 2004 Olympic Games on the Greek economy: A small macroeconometric model

This paper examines the impact of the Athens 2004 Olympic Games on the Greek economy. Using a small aggregate macroeconometric model we find evidence to support the view that the Olympics is an event that could successfully boost the economy of the host city by generating benefits that outweigh the preparation cost. Consistent with recent literature in this area, whilst the impact effects are quite strong during the preparation phase and the year the Games took place, the long-term economic legacy effects appear to be quite modest.     

Empirical determination of aggregate demand and supply curves: The example of the RWI Business Cycle Model

Macroeconometric models are often criticised for being too complex and difficult to read in theoretical terms. To overcome these difficulties, Hickman suggested the calculation of a model's implicit aggregate demand/supply (AD/AS) structure. The method helps to understand models and their main properties in theoretical terms and facilitates detailed model comparisons. This paper uses the AD/AS–IS/LM apparatus to analyse the simulation properties of the RWI (Rheinisch-Westfälisches Institut für Wirtschaftsforschung) Business Cycle Model, a medium-sized short-term macroeconometric model for Germany. The results confirm theoretical expectations for AD and AS elasticities and reveal particular reactions linked to peculiar model specifications such as an endogenous government sector. The results are also much in line with a previous study in a multi-country model context.     

Cost-based Phillips Curve forecasts of inflation

It is a well-established idea that prices are a function of marginal cost, yet estimating a reliable measure of marginal cost is difficult to do. Stock and Watson (1999) use the Phillips Curve to forecast inflation for a variety of existing activity variables that researchers commonly use to proxy for marginal cost. This paper uses a similar type of approach to examine the performance of a new candidate for the activity variable, which is marginal cost measured following the theoretical methodology of Bils (1987), which we find to be simple yet powerful when implemented empirically. We then use the Phillips Curve to conduct pseudo out-of-sample inflation forecasts for the US using: output, unemployment, hours, the labor share, the capacity utilization rate, and the new measure of marginal cost. For almost all cases, forecast errors are lowest in the regressions with the new marginal cost variable, indicating that this new measure is an improvement over previous attempts to proxy for marginal cost.     

Policy analysis in real time using IMF's monetary model

We investigate to what extent estimated relationships of the IMF's monetary model and their policy implications are sample dependent. We observe that estimates of the model's key parameters and model-based measures of macroeconomic disequilibria are highly dependent on the data vintage employed. Changes in parameter estimates solely due to data revisions are found to be much smaller than those owing to parameter instability, possibly reflecting model misspecification. Moreover, instability in parameter estimates contributes to more uncertainty in assessments of macroeconomic disequilibria than data revisions. Analyses based on a version of the model in difference form are also found to be quite sensitive to the data vintage employed, although to a lesser extent than those based on the standard version of the model with variables in levels.     

Unveiling the objectives of central banks: Tales of four Latin American countries

We estimate the stabilization objectives of four Latin American countries that have implemented a flexible inflation targeting regime recently: Brazil, Chile, Colombia and Peru. In doing so, we develop a New Keynesian dynamic stochastic general equilibrium model for these economies and estimate their structural parameters through Bayesian methods. To infer the stabilization objectives in each country, we assume that central banks set monetary policy optimally. Our main results highlight that the central banks in these four countries have a high preference for stabilizing inflation, but do not have the systematic objective of stabilizing the exchange rate. This result is robust to assuming either commitment or discretion in the optimal policy. Also, in contrast to the case of commitment, assuming discretion in the optimal monetary policy increases the preference for interest rate smoothing, making it comparable to a preference for inflation stabilization. Finally, except for the case of Peru, the monetary policy under discretion has a better empirical fit in these countries than the one under commitment.     

The type-agent core for exchange economies with asymmetric information

The type-agent core is a new solution concept for exchange economies with asymmetric information. It coincides with the set of subgame-perfect equilibrium outcomes of a simple competitive screening game. Uninformed intermediaries help the agents to cooperate in an attempt to make some profit. The paper extends the work of Perez-Castrillo [Cooperative outcomes through non-cooperative games, Games Econ. Behav. 7 (1994) 428-440] to exchange economies with non-transferable utility and asymmetric information. The type-agent core is a subset of Wilson's coarse core [Wilson, Information, efficiency, and the core of an economy, Econometrica 46 (1978) 807-816]. It is never empty, even though it may be a strict subset of Wilson's fine core. In addition, it converges towards the set of constrained market equilibria as the economy is replicated.     

Conditional forecasts on SVAR models using the Kalman filter

This note shows how conditional forecasts from identified VAR models can be computed using Kalman filtering techniques. These techniques are nowadays routine for applied macroeconomists, and hence the computation of conditional forecasts using these methods are simple to implement.     

A Bayesian method of combining judgmental and model-based density forecasts

This paper introduces a formal method of combining expert and model density forecasts when the sample of past forecasts is unavailable. It works directly with the expert forecast density and endogenously delivers weights for forecast combination, relying on probability rules only. The empirical part of the paper illustrates how the framework can be applied in forecasting US inflation by mixing density forecasts from an autoregressive model and the Survey of Professional Forecasters.     

Optimal monetary policy and economic growth

A question at the center of many analyses of optimal monetary policy is, why do central banks never implement the Friedman rule? To the list of answers to this question, we add neoclassical production (specifically, the Tobin effect) as one possible explanation. To that end, we study an overlapping generations economy with capital where limited communication and stochastic relocation create an endogenous transactions role for fiat money. We assume a production function with a knowledge externality (Romer style) that nests economies with endogenous growth (AK form) and those with no long-run growth (the Diamond model). The Tobin effect is shown to be always operative. Under CRRA preferences, a mild degree of social increasing returns is sufficient (but not necessary) for some positive inflation to dominate zero inflation and for the Friedman rule to be sub-optimal, irrespective of the degree of risk aversion.     

New models of trader beliefs and their application for explaining financial bubbles

One challenging and exigent problem in behavior finance is how to establish verifiable models describing the appearance and burst of price bubbles. Current results are enhanced in this paper through a series of improvement as follows: new models are proposed for describing the return and dividend processes, especially the trader's behavior with the adaptive expectation belief and the bounded rational expectation belief, respectively; with these models, we establish dynamical systems in terms of the price-to-earnings ratio and the forecast-to-earnings ratio; the detailed solution and asymptotic analysis of these equations provide new, elaborate and quantitative explanations for both the formation and disappearance of different price bubbles; inspired by the herd behavior framework, a new random belief evolutionary mechanism is devised to model the belief change between two beliefs; a specific genetic algorithm is designed to efficiently estimate model parameters; simulation and empirical studies are carried out to illustrate the application of new methods. Both theoretical and empirical results sufficiently show the reasonability, practicality, efficiency and robustness of our new models and methods for properly explaining the appearance and burst of different kinds of price bubbles.     

Fiscal devaluation and economic activity in the EU

In the aftermath of the global financial crisis, a fiscal devaluation (hereafter, FD), understood as a shift in taxation from labour to consumption, has been debated as a possible tool for restoring competitiveness, particularly in peripheral countries of the Eurozone. We contribute to this debate. Based on a set of panel and spatial panel models for the EU 27 over the period 1995–2014, we find that FD works, especially where economic activity is heavily subdued and in sectors more exposed on external competition. FD increases value added in exports, improves net exports, accelerates GDP and employment growth, and decelerates growth in labour costs. These effects are nonlinear; they are stronger in the members of the Eurozone and weaker in countries with either more coordinated or more centralised wage bargaining processes or more generous unemployment benefits. The magnitude of these effects is dampened by strict regulatory barriers: they are weaker in countries with higher barriers to entrepreneurship, trade and investment. Most importantly, FD is not a beggar thy neighbour policy, at least in the EU. In our sample, the aggregate demand (‘cooperative’) effect of unilateral FD, which is beneficial for neighbouring countries, outweighs by far the expenditure switching (‘competitive’) effect, which comes at the expense of other countries’ competitiveness. FD implemented in one country can benefit other countries, provided that they are strongly integrated in global value chains. These findings are robust to changes in the estimation methods, the sample composition, the set of explanatory variables and the selection of a spatial weight matrix.