We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 26 March 2020
The construction of an economic forecast involves a blend of the use of a set of formal equations summarising to the best of the model-builders' ability the dominant characteristics of the past behaviour of the economy, together with a complex set of judgements about the way in which these equations have recently behaved and are likely to behave over the forecast period. We, in common with other forecasters, have published papers about the characteristics of our formal model and about particular equations, but little about the rather flexible way in which the model is actually used to produce a forecast. This Note is an attempt to redress the balance somewhat. It is also intended as an explanatory note to the table of residual adjustments which, beginning with this issue of the Review (see p. 19), we intend to publish regularly as part of the background to our quarterly forecasts.
(1) Occasionally, though in our experience rather rarely, it may be possible to cope with these effects by means of the technique of ‘dummy variables’. These are variables which, for example, take the value 0 or 1 according to whether a particular qualitative influence is held to be operating or not— for instance whether there is an incomes policy. This tech nique can allow for both intercept and slope changes in a relationship (see for example J. Johnston, Econometric Methods, 2nd edition, pp. 176-85). The main difficulty with the technique is that it assumes a uniform quantitative influence of whatever qualitative factor is at work, and this is frequently implausible.
(1) See D. Savage, ‘Interpreting the investment intentions data’, National Institute Economic Review, No. 73, August 1975.
(2) See D. J. O'Dea, Cyclical Indicators for the Postwar British Economy, NIESR Occasional Paper XXVIII, 1975; Economic Trends, ‘Cyclical indicators for the UK Economy’, (Central Statistical Office, monthly; see especially March 1975 and May 1976).
