The past few weeks have seen turmoil and turbulence in global financial markets, with 21 January seeing the sharpest one-day share price falls in many Asian and European markets since 2001, ostensibly in response to widespread fears of an imminent recession in the US. A careful examination of the available real indicators suggests that much of the recent panic in financial markets is not based on fundamentals, although the aggressive response of the Federal Reserve, with the biggest rate move since 1994 at an unscheduled meeting, gave the impression that it was privy to information that had not yet appeared in hard data, such as the imminent collapse of a major US bank. However it is also possible that the Federal Reserve acted precipitately to technical fall out from losses at Société Générale in France, which seems to have sparked much of the panic trading. The Banque de France informed the Federal Reserve of the matter in advance of their meeting scheduled for the following week. It is possible that if the Federal Reserve had waited for all the information they needed they might not have acted, and indeed they may have damaged their credibility by their precipitate action.

Between NIESR's October 2007 and January 2008 global economy forecasts, oil prices rose by almost $20 per barrel, and they averaged $86 a barrel in the fourth quarter of 2007, as against our assumption of $71 a barrel. In the first quarter of 2008 we project that they will average $90 a barrel, $19 above the level we projected in October. Real oil prices have risen almost to the level seen in around 1980, as we can see from figure 1. There has, however, been no resurgence of inflationary pressure on the scale that was observed in the mid-1970s and early 1980s in response to oil price increases. This has been much as we anticipated, as is discussed in Barrell and Pomerantz (2004), and our model simulation results reflect the changed environment in which the increase has taken place.

Despite the turmoil in global financial markets during the second half of 2007, UK GDP continued to grow at a robust pace. The above trend growth of 2006 continued into the first three quarters of last year and the preliminary estimate of GDP growth shows only a limited moderation into the fourth quarter of last year (figure 1). This suggests we will see a growth rate of 3.1 per cent for 2007 as a whole, a slight acceleration from 2.9 per cent in 2006. This robust growth has been sustained by household consumption and, to a lesser extent, gross fixed investment.

Sterling has fallen markedly against other currencies in the past few months, and in January reached its lowest point against the euro (or its equivalent) since the last quarter of 1996, as we can see from figure 1, which uses the first three weeks of January 2008 as an indicator of the value that will be achieved in the whole first quarter. Sterling has also weakened against the dollar in the past few months, although it remains at a high level. In effective terms, sterling fell on average by 2.5 per cent in the fourth quarter, and in early January it was more than 6 per cent lower than the average for the previous quarter. The fall was largely unanticipated, and the effective exchange rate for the first quarter of 2008 is more than 7 per cent below where we assumed it would be in October 2007.

This paper evaluates the probability density forecasts reflected in the Bank of England's real GDP growth fan charts. Evaluation is carried out using tests that allow for data dependence and using two GDP growth estimates. Results suggest there are problems with the shorter horizon forecasts, but conclusions about the performance of longer-term forecasts depend to some extent on the GDP estimates used in the assessment.

This article analyses the forecasts of inflation and GDP growth supplied by the individual respondents to the Bank of England's quarterly Survey of External Forecasters, 1996-2005, using a recently released dataset. This comprises a conventional incomplete panel dataset, with an additional dimension arising from the collection of repeated forecasts for the last quarter of each year. This fixed-event forecast structure allows study of the forecast revision process, its weak and strong efficiency, and its relation to macroeconomic news. The collection of density forecasts as well as point forecasts allows study of the revision process of forecast uncertainty.

An overview is provided of the issues raised in the recent literature on the use of real-time data in the context of nowcasting and forecasting UK macroeconomic events. The ideas are illustrated through two specific applications using UK real-time data available over 1961-2006 and providing probability forecasts that could have been produced in real time over the past twenty years. In the first, we consider the reliability of first-release data on the components of UK aggregate demand by looking at forecasts of the probability of substantial data revisions. In the second, we consider the estimation of the output gap, illustrating the uncertainty surrounding its measurement through density forecasts and focusing on its interpretation in terms of inflationary pressure through an event probability forecast.

This paper uses vector error correction models of Switzerland for forecasting output, inflation and the short-term interest rate. It considers three different ways of dealing with forecast uncertainties. First, it investigates the effect on forecasting performance of averaging over forecasts from different models. Second, it considers averaging forecasts from different estimation windows. It is found that averaging over estimation windows is at least as effective as averaging over different models and both complement each other. Third, it examines whether using weighting schemes from the machine learning literature improves the average forecast. Compared to equal weights the effect of alternative weighting schemes on forecast accuracy is small in the present application.

This paper aims to provide a brief and relatively non-technical overview of state-of-the-art forecasting with large data sets. We classify existing methods into four groups depending on whether data sets are used wholly or partly, whether a single model or multiple models are used and whether a small subset or the whole data set is being forecast. In particular, we provide brief descriptions of the methods and short recommendations where appropriate, without going into detailed discussions of their merits or demerits.