Economic forecasts are usually presented as point estimates, despite the margin of uncertainty which surrounds them. In
November 1995 the National Institute began to present estimates of the probability of the government's inflation target being met
and of there being a fall in GDP. This article describes the methods that we use for calculating these probabilities. We show, by
studying eight successive forecasts of the same event, how forecast reliability improves as the forecast horizon approaches and
demonstrate that this can be explained in terms of the accumulation of information about the state of the economy.
Get full access to this article
View all access options for this article.
References
1.
Blake, A. (1996), 'Forecast error bounds by stochastic simulation' , National Institute Economic Review, no. 156, pp..
2.
Britton, A. and Pain, N. (1992), Economic Forecasting in Britain, National Institute Report no. 4.
3.
Clements, M. (1995), 'Rationality and the role of judgement in macroeconomic forecasting', EconomicJournal, 105, pp. 510-20.
4.
Nordhaus, W. (1987), 'Forecasting efficiency: concepts and applications', Review ofEconomics and Statistics, vol. 69, pp. 714-35.
5.
Pain, N. (1994), 'Cointegration and forecast evaluation: some lessons from National Institute forecasts', Journal of Forecasting , no. 13, pp. 481-493.
6.
Sefton, J. and Weale, M. (1995), Reconciliation of National Income and Expenditure: Balanced National Accounts for the United Kingdom, 1920-1990, Cambridge University Press, Cambridge.
7.
Wallis, K. (1989), 'Macroeconomic forecasting: a survey', Economic Journal, no. 99, pp. 28-61.
8.
Wallis, K. and Whitley, J. (1991), 'Sources of error in forecasts and expectations: UK economic models, 1984-88', Journal of Forecasting, no. 10, pp. 231-53.
