Showing papers on "Predictability published in 1995"

Exchange rates and fundamentals: Evidence on long-horizon predictability

Abstract: Regressions of multiple-period changes in the log exchange rate on the deviation of the log exchange rate from its 'fundamental value' display evidence that long-horizon changes in log nominal exchange rates contain an economically significant predictable component. To account for small-sample bias and size distortion in asymptotic tests, inference is drawn from bootstrap distributions generated under the null hypothesis that the log exchange rate is unpredictable. The bias-adjusted slope coefficients and R[superscript]2's increase with the forecast horizon, and the out-of-sample point predictions generally outperform the driftless random walk at the longer horizons. Copyright 1995 by American Economic Association.

Predictability of Stock Returns: Robustness and Economic Significance

Abstract: This article examines the robustness of the evidence on predictability of U.S. stock returns, and addresses the issue of whether this predictability could have been historically exploited by investors to earn profits in excess of a buy-and-hold strategy in the market index. We find that the predictive power of various economic factors over stock returns changes through time and tends to vary with the volatility of returns. The degree to which stock returns were predictable seemed quite low during the relatively calm markets in the 1960s, but increased to a level where, net of transaction costs, it could have been exploited by investors in the volatile markets of the 1970s. MANY RECENT STUDIES CONCLUDE that stock returns can be predicted by means of publicly available information, such as time series data on financial and macroeconomic variables with an important business cycle component.' This conclusion seems to hold across international stock markets as well as over different time horizons. Variables identified by these studies to have been statistically important for predicting stock returns include interest rates, monetary growth rates, changes in industrial production, inflation rates, earnings-price ratios, and dividend yields. However, the economic interpretation of these results is controversial and far from evident. First, it is possible that the predictable components in stock returns reflect time-varying expected returns, in which case predictability of stock returns is, in principle, consistent with an efficient stock market. A second interpretation takes expected returns as roughly constant and regards predictability of stock returns as evidence of stock market inefficiency. It is, however, clear that predictability of excess returns on its own does not imply stock market inefficiency, and can be

On the Predictability of Stock Returns: An Asset-Allocation Perspective

Abstract: Sample evidence about the predictability of monthly stock returns is considered from the perspective of an investor allocating funds between stocks and cash. A regression of stock returns on a set of predictive variables might seem weak when described by usual statistical measures, but such measures can fail to convey the economic significance of the sample evidence when it is used by a risk-averse Bayesian investor to update prior beliefs about the regression relation and to compute an optimal asset allocation. Even when those prior beliefs are weighted substantially against predictability, the current values of the predictive variables can exert a strong influence on the portfolio decision.

An Improved Procedure for EI Niño Forecasting: Implications for Predictability

Abstract: A coupled ocean-atmosphere data assimilation procedure yields improved forecasts of El Nino for the 1980s compared with previous forecasting procedures. As in earlier forecasts with the same model, no oceanic data were used, and only wind information was assimilated. The improvement is attributed to the explicit consideration of air-sea interaction in the initialization. These results suggest that EI Nino is more predictable than previously estimated, but that predictability may vary on decadal or longer time scales. This procedure also eliminates the well-known spring barrier to EI Nino prediction, which implies that it may not be intrinsic to the real climate system.

Do Arbitrage Pricing Models Explain the Predictability of Stock Returns

Abstract: This article studies predictability in U.S. stock returns for multiple investment horizons. The authors measure to what extent predictability is driven by premiums for economywide risk factors, comparing two standard methods for factor selection. They study single-beta models and multiple-beta models. The authors show how to estimate the fraction of the predictability in returns captured by the model simultaneously with the other parameters. Their analysis indicates that the models capture a large fraction of the predictability for all of the investment horizons. The performance of the principal components and the prespecified-factor approaches are broadly similar. Copyright 1995 by University of Chicago Press.

Time-Varying Expected Returns in International Bond Markets

Abstract: This article examines the predictable variation in long-maturity government bond returns in six countries. A small set of global instruments can forecast 4 to 12 percent of monthly variation in excess bond returns. The predictable variation is statistically and economically significant. Moreover, expected excess bond returns are highly correlated across countries. A model with one global risk factor and constant conditional betas can explain international bond return predictability if the risk factor is proxied by the world excess bond return, but not if it is proxied by the world excess stock return. A GROWING BODY OF LITERATURE describes predictable variation in U.S. and international asset returns. There are two competing views regarding the source of the predictability. Some authors interpret the return predictability as evidence of market inefficiency, while others attribute it to rational variation in required asset returns. In this article, I study predictable variation in international government bond returns, focusing on two questions: (1) Can the excess returns of long-term bonds be forecast using global or country-specific instruments? and (2) Is the observed behavior of expected excess bond returns consistent with a simple asset pricing model, market efficiency, and international market integration? Studying this subset of capital markets is useful because bond returns are affected by relatively few factors. The excess returns of long-term government bonds' are subject only to interest rate risk. There is no default risk or cash flow uncertainty, and almost all foreign exchange risk can be hedged. The simplicity of government bonds facilitates the identification of useful forecasting instruments and the interpretation of empirical findings. Assuming ra

Implementing Option Pricing Models When Asset Returns Are Predictable

Abstract: The predictability of an asset's returns will affect the prices of options on that asset, even though predictability is typically induced by the drift, which does not enter the option pricing formula. For discretely-sampled data, predictability is linked to the parameters that do enter the option pricing formula. We construct an adjustment for predictability to the Black-Scholes formula and show that this adjustment can be important even for small levels of predictability, especially for longer maturity options. We propose several continuous-time linear diffusion processes that can capture broader forms of predictability, and provide numerical examples that illustrate their importance for pricing options. THERE IS NOW A substantial body of evidence that documents the predictability of financial asset returns.' Despite the lack of consensus as to the sources of such predictability-some attribute it to time-varying expected returns, perhaps due to changes in business conditions, while others argue that predictability is a symptom of inefficient markets or irrational investors-there is a growing consensus that predictability is a genuine feature of many financial asset returns. In this article, we investigate the impact of asset return predictability on the prices of an asset's options. A comparison between the polar cases of perfect predictability (certainty) and perfect unpredictability (the random walk) suggests that predictability must have an effect on option prices, although what that effect might be is far from obvious. However, in the

Error growth and estimates of predictability from the ECMWF forecasting system

Abstract: Examination has been made of the skill of ECMWF forecasts of the 500 hPa height field produced daily out to ten days ahead, verifying in the period from 1 December 1980 to 31 May 1994. Over this time accuracy has been improved substantially over the first half of the forecast range. the systematic (seasonal-mean) component of the error has been greatly reduced at all forecast times, but there has been little reduction in the non-systematic (transient) component later in the range. The simple model proposed by Lorenz for the intrinsic growth of forecast error has been applied to the evolution of differences between consecutive forecasts. the implied growth-rates of small forecast errors have increased significantly since 1981. They do not show much variation with season, and are a little lower in the southern than in the northern hemisphere. the most recent error-doubling times are around 1.5 days for the northern hemisphere and 1.7 days for the southern hemisphere. Error saturation levels are at present similar to or greater than those of the 1981 version of the model, having been significantly lower in intermediate years. the accuracy of recent short- and early medium-range forecasts and realism of the climatology of the forecast model support the view that estimates of intrinsic error-growth parameters from the current forecasting system are more reliable than those obtained earlier. Forecast accuracy later in the medium range may thus not have benefited fully from improvements earlier in the range because of the faster error-growth associated with a more active, though more realistic, forecast model. Overprediction of variance may nevertheless detrimentally affect present levels of skill and estimates of predictability in all seasons other than summer. The error-growth model currently indicates that it is possible, in principle, to make deterministic mediumrange forecasts for the extratropical 500 hPa height field of the northern hemisphere that are as accurate five days ahead as present forecasts are three days ahead, provided the one-day forecast error can be reduced by the same factor in the future as has actually been achieved in the years since 1981. the level of error currently reached at day seven would then be reached at around day ten. the scope for improvement of forecasts for the southern hemisphere appears to be rather larger. Improvements seem to be possible throughout the spectral range studied, up to total wave-number 40. This is found also for the rotational and divergent wind components at 850 and 200 hPa. For these components, particularly the divergent component, there is a quite pronounced error in the representation of the largest scales.

The annual cycle and the predictability of the tropical coupled ocean-atmosphere system

Abstract: Using large-scale circulation statistics from the Pacific Ocean basin, predictability of the coupled ocean-atmosphere system on interannual time scales is found both to be limited in extent and to possess a strong annual cycle. Irrespective of when lagged correlations are commenced, correlations decrease rapidly through the boreal spring, indicating an inherent predictability limitation for large scale coupled oceanicatmospheric processes such as El Nino. Long term prediction experiments using numerical coupled-models show that the models are excellent facsimiles of the real system. They, too, encounter the predictability barrier and exhibit a substantial decrease in observation-prediction correlation across the boreal spring. Thus, a predictive system based solely on the interactive physics of the Pacific Basin appears limited to a maximum of less than one year and a minimum of only one or two months. Two hypotheses are made to explain the existence of the predictability barrier. First, it is argued that the tropical coupled system is at its frailest state during the boreal spring and that the signal-to-noise ratio is weakest. In such a system, maximum random error growth may occur as the atmosphere and the ocean become temporally detached and wander onto different climate trajectories. A series of 144 preliminary Monte Carlo experiments were conducted with a coupled ocean-atmosphere model to test the hypothesis. Irrespective of when the experiments were commenced, error growth was maximized at the same time of the year. The second hypothesis suggests that the near-equatorial circulation is perturbed at the time of its weakest state by external influences such as the monsoon and that the climate wanderings are “nudged” deterministically. There is observational and theoretical evidence to support the hypothesis. Observations suggest that anomalous monsoons impart basin-wide coherent alterations of the wind stress field in the Pacific Ocean. Experiments with a coupled ocean-atmosphere model show that the period of an ENSO event is altered substantially by an anomalous monsoon. Given that there appear to be precursors to anomalous monsoons, it is suggested that there may be ways to avoid the predictability barrier and thus extend prediction of the entire system. Finally, noting that the two hypotheses are not mutually exclusive, they are combined to form a unified theory. As the asymmetric monsoonal and the symmetric near-equatorial heating are in approximate quadrature, it is argued that the monsoons influence the Walker circulation during the boreal spring. However, during the boreal fall and early winter the near-equatorial heating variability dominates the winter monsoon.

Feasibility of Seasonal Forecasts Inferred from Multiple GCM Simulations

Abstract: Assuming that SST provides the major lower boundary forcing for the atmosphere, observed SSTs are prescribed for an ensemble of atmospheric general circulation model (GCM) simulations. The ensemble consists of 9 “decadal” runs with different initial conditions chosen between 1 January 1979 and 1 January 1981 and integrated about 10 years. The main objective is to explore the feasibility of seasonal forecasts using GCMS. The extent to which the individual members of the ensemble reproduce the solutions of each other (i.e., reproducibility) may be taken as an indication of potential predictability. In addition, the ability of a particular GCM to produce realistic solutions, when compared with observation, must also be addressed as part of the predictability problem. A measure of reproducibility may be assessed from the spread among ensemble members. A normalized spread index, σn/σs, can be defined at any point in space and time, as the variability of the ensemble (σn) normalized by the climatologic...

Mesoscale Predictability and the Spectrum of Optimal Perturbations

Abstract: The spectrum of finite-time most unstable structures, also referred to as singular vectors (SVs), is computed for a regional, mesoscale primitive-equation model. The number of growing SVs present in this spectrum is of interest for investigating mesoscale predictability since it provides an estimate of the dimension of the unstable subspace of the model phase space. This dimension is used to critically assess the contrasting conclusions that have been reached by different authors in mesoscale predictability studies. Computations are carried out for two different synoptic cases (explosive cyclogenesis over the North Atlantic and Alpine lee cyclogenesis) using two different norms. The first is loosely related to total perturbation energy and the second measures the energy of rotational normal modes only. The latter is designed to reduce the influence of geostrophic adjustment on the measure of growth. The models used are the tangent-linear and adjoint components of the dry-adiabatic version of the ...

Maximizing Predictability in the Stock and Bond Markets

Abstract: We construct portfolios of stocks and of bonds that are maximally predictable with respect to a set of ex ante observable economic variables, and show that these levels of predictability are statistically significant, even after controlling for data-snooping biases. We disaggregate the sources for predictability by using several asset groups, including industry-sorted portfolios, and find that the sources of maximal predictability shift considerably across asset classes and sectors as the return-horizon changes. Using three out-of-sample measures of predictability, we show that the predictability of the maximally predictable portfolio is genuine and economically significant.

Error Growth in a Large Numerical Forecast System

Abstract: Internal and external z500 global total rms errors followed quadratic growth laws quite well in NMC Medium-Range Forecast (MRF) Model 0–10-day forecasts for 1988–93. Growth parameters and model and analysis errors for many winter were estimated using the quadratic rms error growth assumption. Both the MRF model error and analysis error have nearly halved during 1988–93. But at the same time the growth parameters have nearly doubled: smaller errors grow faster. Thus while the limit of deterministic predictability (rms error 71% of saturation) has been going up, the limit of dynamic predictability (rms error 97.5% of saturation) seems to be set at around 20 days in large horizontal scales, dropping to 6–7 days in small scales.

Transient error growth and local predictability: a study in the Lorenz system

Abstract: Lorenz's three-variable convective model is used as a prototypical chaotic system in order to develop concepts related to finite time local predictability. Local predictability measures can be represented by global measures only if the instability properties of the attractor are homogeneous in phase space. More precisely, there are two sources of variability of predictability in chaotic attractors. The first depends on the direction of the initial error vector, and its dependence is limited to an initial transient period. If the attractor has homogeneous predictability properties, this is the only source of variability of error growth rate and, after the transient has elapsed, all initial perturbations grow at the same rate, given by the first (global) Lyapunov exponent. The second is related to the local instability properties in phase space. If the predictability properties of the attractor are not homogeneous, this additional source of variability affects both the transient and post-transient phases of error growth. After the transient phase all initial perturbations of a particular initial condition grow at the same rate, given in this case by the first local Lyapunov exponent. We consider various currently used indexes to quantify finite time local predictability. The probability distributions of the different indexes are examined during and after the transient phase. By comparing their statistics it is possible to discriminate the relative importance of the two sources of variability of predictability and to determine the most appropriate measure of predictability for a given forecast time. It is found that a necessary premise for choosing a relevant local predictability index for a specific system is the study of the characteristics of its transient. The consequences for the problem of forecasting forecast skill in operational models are discussed. DOI: 10.1034/j.1600-0870.1995.00006.x

Predictability Bias in the U.S. Equity Market

Abstract: In an investigation of an asset pricing anomaly, “predictability bias,” stocks were classified into quintiles based on how predictable the company's earnings have been in the past, as measured by analysts' past forecast errors. “Current” forecasts of earnings are excessively optimistic for companies whose earnings were hard to predict in the past; that is, the least predictable companies have much larger, positive forecast errors (forecast > actual) relative to the most predictable companies. Abnormal returns are consistent with the current forecast errors; that is, stocks of the least predictable companies substantially underperform stocks of the most predictable companies. Adjustments for systematic risk, firm size, book-to-price ratio, and industry factors do not eliminate the differential returns between least predictable and most predictable companies.

ENSO and ENSO-Related Predictability. Part II: Northern Hemisphere 700-mb Height Predictions Based on a Hybrid Coupled ENSO Model

Abstract: Long-range sea surface temperature forecasts from two different coupled ocean-atmosphere models of the tropical Pacific are used in conjunction with statistical models relating winter Northern Hemisphere 700-mb height and tropical SST to forecast the former field at a lead time of two seasons in advance. The forecasts show considerable skill over large areas, with a regional distribution of predictive performance that is consistent with the observed contemporaneous relation between the two fields. Comparable skills for lead time of a year or more in advance seem likely.

Variability of Marine Fog Along the California Coast

Abstract: Author(s): Filonczuk, Maria K; Cayan, Daniel R; Riddle, Laurence G | Abstract: Visibility in coastal regions has a significant impact on government, commercial, and private sector activities. The primary phenomenon significantly affecting visibility along the western United States coastal regions is fog. Fog is a natural hazard to boating, commercial shipping, and other waterway activities.The West Coast of the United States has been identified as one of the major fog producing regions of the world. Present accuracy in predicting marine coastal fog and low stratus clouds is limited. Although most weather forecasting has improved with recent advances in atmospheric circulation models and satellite observations, there is relatively little operational guidance for the prediction of marine and coastal fog.Currently the forecasting of visibility relies mainly on the availability of local observations and experience with local weather tendencies. National Weather Service and other forecasters typically focus on selected local conditions which usually presage fog formation and if a sufficient number are present, they will forecast possible fog for the region. However, this level of information and experience is not available in all coastal areas. Results in this study suggest that attention to the large-scale circulation in addition to local conditions may lead to increased skill and accuracy in extended range forecasts.The large-scale structure and interannual variability of fog have not been well described because data has been unavailable and a large volume of data is required. In the present analysis, several years of observations from a set of coastal weather stations and a set of comprehensive marine weather reports (primarily ships) is used to examine large-scale processes and local conditions which affect fog formation. A discussion of fog observation criteria is given. In addition, evidence is provided by a collection of observations of hours of fog-horn operations at several coastal sites over 15 years, starting in 1950.The relationship between local and large-scale conditions, and fog and stratus formation may yield improved predictability of fog. This study indicates that large-scale conditions, especially regional-hemispheric circulation, are a vital component of both high and low fog occurrences; moreover, atmospheric circulation in the Eastern North Pacific Ocean and on the West Coast may contribute to improved forecasts for the development and persistence of fog.

Local Lyapunov exponents: Predictability depends on where you are

Abstract: The dominant Lyapunov exponent of a dynamical system measures the average rate at which nearby trajectories of a system diverge. Even though a positive exponent provides evidence for chaotic dynamics and upredictability, there may predictability of the time series over some finite time periods. In this paper one version of a local Lyapunov exponent is defined for a dynamic system perturbed by noise. These local Lyapunov exponents are used to detect the parts of the time series that may be more predictable than others. An examination of the fluctuations of the local Lyapunov exponents about the average exponent may provide important information in understanding the heterogeneity of a system. We will discuss the theoretical properties of these local exponents and propose a method of estimating these quantities using nonparametric regression. Also we will present an application of local exponents for interpreting surface pressure data.

Chapter 17 On the predictability of common stock returns: World-wide evidence

Abstract: Publisher Summary Research in finance over the past 10 to 15 years has revealed stock price behavior that is inconsistent with the predictions of familiar models. This chapter discusses recent empirical findings on the predictability of stock returns. These findings document persistent cross-sectional and time-series patterns in returns that are not predicted by extant theory. As a result, such empirical regularities are often classified as anomalies. The nature of the evidence discussed is interpreted by many market observers as convincing evidence of market inefficiency The joint null hypothesis underlying such analyses is that security markets are efficient and returns behave according to a prespecified equilibrium model. If the joint hypothesis is rejected, that rejection cannot be specifically attributed to one or the other branch of the hypothesis– a conclusion that securities markets are not efficient is inapproriate since the rejection may be because of a test design based on an incorrect equilibrium model. While it is acknowledge that the strict form of market efficiency discussed in most analysis is an unlikely description of security price determination, the simple fact that so many of these regularities have persisted for more than 50 years suggests that perhaps the benchmark models are less than complete descriptions of equilibrium price formation.

Forecasting nonlinear economic time series: A simple test to accompany the nearest neighbor approach

Abstract: This paper is based on a recent nonparametric forecasting approach by Sugihara, Grenfell and May (1990) to improve the short term prediction of nonlinear chaotic processes. The idea underlying their forecasting algorithm is as follows: For a nonlinear low-dimensional process, a state space reconstruction of the observed time series exhibits “spatial” correlation, which can be exploited to improveshort term forecasts by means of locally linear approximations. Still, the important question of evaluating the forecast perfomance is very much an open one, if the researcher is confronted with data that are additionally disturbed by stochastic noise. To account for this problem, a simple nonparametric test to accompany the algorithm is suggested here. To demonstrate its practical use, the methodology is applied to observed price series from commodity markets. It can be shown that the short term predictability of the best fitting linear model can be improved upon significantly by this method.

Simulation of a long-lived meso-? scale convective system over the mediterranean coast of spain. Part I: Numerical predictability

Abstract: An unusually long-lived (33 hours), devastating (local maximum rainfall rate over 800 mm/24 hr) meso-β-scale (diameter smaller than 200 km) convective system that occurred over the Mediterranean coast of Spain has been simulated reasonably well by means of a regional numerical model Several runs of the model including parameterized convection and boundary conditions of varying degrees of complexity have been performed In most of these experiments, the main characteristics of the event, namely its, stationarity and duration, are captured The direct relationship between the Lagrangian lifetime of a meteorological system and its degree of deturministic predictability seems to be corroborated by the results: It appears that the meso-α-scale forcing that preceded and favoured the MCS was especially well predictable, and once initiated, the simulated MCS seems to have several feedback mechanisms helping to extend its life Results are encouraging, because they reveal that it might be possible to predict very severe episodes of small MCSs such as the one shown here sufficiently in advance

Multiresolution methods for financial time series prediction

Abstract: Summary form only given. Fractional Brownian motion (fBm), a 1/f, fractal process, has long been considered a plausible model for financial time series. A fractal structure of the market, indicating the presence of correlations across time, hints at the possibility of some predictability. Recent advances in time/frequency localized transforms by the applied mathematics and electrical engineering communities provide us with powerful new methods for the analysis of this type of process. In fact, it has been proven by Wornell that the wavelet transform is an optimal (KL) transform for fBm processes. With this result, we consider using the wavelet decomposition to analyze financial time series. Specifically, the discrete wavelet transform can be used to decompose a signal into several scales, while maintaining time localization of events in each scale. In terms of financial time series, we can conceptually think of each of these scales as the contribution to the price movement from the information and traders associated with a given investment horizon, for instance, long term traders, such as institutional investors, basing their trades on long term information, form the low-frequency component of the market. Once we have extracted out these scales, we can view each as a stationary time series, which can be modeled, analyzed and predicted individually, either independently, or in conjunction with other scales and data that is relevant to that scale. For the case of prediction, the forecasts from each scale can be fused together, with traditional techniques such as hard coded decision rules, or with a neural network, to arrive at tomorrow's direction and/or price.

Limits of Predictability of Solar Activity

Abstract: The study of a nonlinear chaotic map of 11-year cycle maxima evolution recently derived from observations is presented with the purpose of predicting the features of the long-term variability of solar activity. It is stressed that dynamical forecast is limited by the Lyapunov time and a statistical approach can be justified due to the ergodic properties of the chaotic evolution. The Gleissberg variation is described as a chaotic walk and its distribution over length is shown to be broad. The global minima are identified as laminar slots of temporal intermittency and their typical distribution over length is also given. We note that a long sunspot cycle can be used as a precursor of the global minimum and a close sequence of global minima (once in approximately 1500–2000 years) may be responsible for the climatic changes (Little Ice Ages).

Numerical weather prediction

Abstract: This review highlights a number of current areas of emphasis in research and operational numerical weather prediction. Detailed accounts of each area of activity are not presented; some key references are provided within each section for interested readers who may wish to explore further. The review outlines the types of weather prediction models where the biggest contributions have emerged in recent years. The topics include an outline of such models, the data, their assimilation and initialization issues, model sensitivity to physical processes, tropical forecast advancements-mon­ soons and hurricanes, newer areas of thrust-use of adaptive grids for mesoscale forecasts of severe weather, and finally the recognition of current limits of predictability-chaos, i.e. the need for ensemble forecasts, via the probabilistic Monte Carlo type approach. We have currently reached roughly the halfway point towards the theor­ etical bound of predictability of two weeks, originally stated by Lorenz ( 1 963), for predicting the future state of the atmosphere using large-scale numerical weather prediction models. This limit is currently measured from correlations of observed versus predicted (massor motion-based) fields (generally from 200N to the North pole). The measure of useful skill has slowly increased from 2 days to roughly 7 days in the past 30 years as computing, modeling, and observational strategies have improved. Presently, the predictive skill over the southern hemisphere is roughly half

Operational predictability of winter blocking at ECMWF: an update

Abstract: . Seven winters of analyses and forecasts from the operational archives of the European Centre for Medium Range Weather Forecast had been previously analyzed to assess the performance of the model in forecasting blocking events. This work updates some of this previous diagnostic work to the last five winters, from 1987/88 to 1991/92. The data set therefore covers all winter seasons (DJF) from 1980/81 to 1991/92, and consists of daily northern hemisphere 500 hPa geopotential height analyses and of the ten corresponding forecasts verifying on the same day ("Lorenz data"). Local blocking and sector blocking have been defined, using different modifications of the original Lejenas and Okland index. The comparison between the first seven and the last five winters, within the restrictions imposed by limited length of the data set, suggests a much improved situation as far as model climatology of blocking is concerned, especially over the Euro-Atlantic region. Operational predictability of blocking as an initial value problem is also shown to be measurably improved, in both Atlantic and Pacific sectors. All such improvements are shown to have taken place together with a considerable reduction of the model systematic error. Nevertheless, forecasting blocking in the medium range remains a difficult task for the model. More work is needed to understand whether the improvements are to be ascribed to the increased model resolution or to better physical parametrisations.