Complete subset regressions

This paper proposes a new method for combining forecasts based on complete subset regressions. For a given set of potential predictor variables we combine forecasts from all possible linear regression models that keep the number of predictors fixed. We explore how the choice of model complexity, as measured by the number of included predictor variables, can be used to trade off the bias and variance of the forecast errors, generating a setup akin to the efficient frontier known from modern portfolio theory. In an application to predictability of stock returns, we find that combinations of subset regressions can produce more accurate forecasts than conventional approaches based on equal-weighted forecasts (which fail to account for the dimensionality of the underlying models), combinations of univariate forecasts, or forecasts generated by methods such as bagging, ridge regression or Bayesian Model Averaging.     

Too similar to combine? On negative weights in forecast combination

This paper provides the first thorough investigation of the negative weights that can emerge when combining forecasts. The usual practice in the literature is to consider only convex combinations and ignore or trim negative weights, i.e., set them to zero. This default strategy has its merits, but it is not optimal. We study the problem from various angles, and the main conclusion is that negative weights emerge when highly correlated forecasts with similar variances are combined. In this situation, the estimated weights have large variances, and trimming reduces the variance of the weights and improves the combined forecast. The threshold of zero is arbitrary and can be improved. We propose an optimal trimming threshold, i.e., an additional tuning parameter to improve forecasting performance. The effects of optimal trimming are demonstrated in simulations. In the empirical example using the European Central Bank Survey of Professional Forecasters, we find that the new strategy performs exceptionally well and can deliver improvements of more than 10% for inflation, up to 20% for GDP growth, and more than 20% for unemployment forecasts relative to the equal-weight benchmark.     

Combining probability forecasts

We consider different methods for combining probability forecasts. In empirical exercises, the data generating process of the forecasts and the event being forecast is not known, and therefore the optimal form of combination will also be unknown. We consider the properties of various combination schemes for a number of plausible data generating processes, and indicate which types of combinations are likely to be useful. We also show that whether forecast encompassing is found to hold between two rival sets of forecasts or not may depend on the type of combination adopted. The relative performances of the different combination methods are illustrated, with an application to predicting recession probabilities using leading indicators.     

Multifactor risk loadings and abnormal returns under uncertainty and learning

We explore the time variation of factor loadings and abnormal returns in the context of a four-factor model. Our methodology, based on an application of the Kalman filter and on endogenous uncertainty, overcomes several limitations of competing approaches used in the literature. Besides taking learning into account, it does not rely on any conditioning information, and it only imposes minimal assumptions on the time variation of the parameters. Our estimates capture both short- and long-term fluctuations of risk loadings and abnormal returns, also showing marked variation across US industry portfolios. The results from mean-variance spanning tests indicate that our baseline model yields accurate predictions and can therefore improve pricing and performance measurement.     

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.     

Combining historical and preliminary information to obtain timely time series data

A method is presented to improve the precision of timely data, which are published when final data are not yet available. Explicit statistical formulae, equivalent to Kalman filtering, are derived to combine historical with preliminary information. The application of these formulae is validated by the data, through a statistical test of compatibility between sources of information. A measure of the share of precision of each source of information is also derived. An empirical example with Mexican economic data serves to illustrate the procedure.     

ECONOMETRIC APPROACHES TO EMPIRICAL MODELS OF EXCHANGE RATE DETERMINATION

Abstract. This paper identifies four principal econometric approaches to the estimation and testing of asset market models of exchange rate determination: the traditional, static reduced-form approach; the error correction and co-integration, dynamic reduced-form approaches; the simultaneous equations approach; and large scale, multi-equation macroeconometric simulation models. Each of these econometric approaches is evaluated with respect to its theoretical validity and the comparative properties of the empirical results obtained. This leads to the conclusion that although there may be little to choose between the different theoretical exchange rate models, there may be grounds for favouring a multi-equation, simultaneous estimation procedure for this class of models.     

Forecasting UK GDP growth and inflation under structural change. A comparison of models with time-varying parameters

Evidence from a large and growing body of empirical literature strongly suggests that there have been changes in the inflation and output dynamics in the United Kingdom. The majority of these papers base their results on a class of econometric models that allows for time-variation in the coefficients and volatilities of shocks. While these models have been used extensively for studying evolving dynamics and for structural analysis, there has been little evidence that they are useful for forecasting UK output growth and inflation. This paper attempts to fill this gap by comparing the performances of a wide range of time-varying parameter models in forecasting output growth and inflation. We find that allowing for time-varying parameters can lead to large and statistically significant gains in forecast accuracy.     

股市收益波动的非对称性研究

随着我国股票市场的逐步发展及证券市场监管体制的逐步完善,股票市场收益波动所呈现的非对称性效应是否会发生逆转引人关注。本文运用不同的GARCH类条件异方差模型(EGARCH、TARCH、PARCH以及非对称CARCH模型),选取1998年1月5日至2006年11月6日上证综指的日收盘价格指数数据,分析和验证了股票收益波动的非对称性效应。结果表明,上海股市股票收益波动仍然具有非对称性,即利空消息对收益波动的冲击效应大于同等程度的利好消息,从而验证了非对称效应的不可逆转性。     

DEMAND,CUSTOMER BASE‐LINE AND DEMAND RESPONSE IN THE ELECTRICITY MARKET: A SURVEY

The increasing use of demand‐side management, as a tool to reliably meet electricity demands at peak time, has stimulated interest among researchers, consumers and producer organiza‐tions, managers, regulators and policymakers. This research reviews the growing literature on models which are used to study demand, customer base‐line (CBL) and demand response in the electricity market. After characterizing the general demand models, the CBL, based on which the demand response models are studied, is reviewed. Given the experience gained from the review and existing conditions, the study combines an appropriate model for each case for a possible application to the electricity market; moreover, it discusses the implications of the results. In the literature, these aspects are studied independently. The main contribution of this survey is attributed to the treatment of the three issues as sequentially interdependent. The review is expected to enhance the understanding of the demand, CBL and demand response in the electricity market and their relationships. The objective is conducted through a combination of demand and supply side managements in order to reduce demand through different demand response programs during peak times. This enables electricity suppliers to save costly electricity generation and at the same time reduce energy vulnerability.