Daily electricity consumption data, available almost in real time, can be used in Italy to estimate the level of industrial production in any given month before the month is over. We present a number of procedures that d o this using electricity consumption in the first 14 days of the month. (This is an extension of a previous model that used monthly electricity data.) We show that, with a number of adjustments, a model using halfmonthly electricity data generates acceptable estimates of the monthly production index. More precisely, these estimates are more accurate than univariate forecasts but less accurate than estimates based on monthly electricity data. A further improvement can be obtained by combining 'halfmonthly' electricity-based estimates with univariate forecasts. We also present quarterly estimates and discuss confidence intervals for various types of forecasts.
KEY WORDS Industrial production Estimation Time series methods Forecasts Leading indicators Infra-monthly series
A number of different methods can be used to generate short-term forecasts of the economic cycle; these include univariate models, models using business surveys, and structural econometric models. In a previous paper (Bodo and Signorini, 1987) we presented a procedure based on electricity consumption data that enabled us to generate reliable estimates of the ltalian industrial production index for month t at the beginning of the following month. Here we extend the methodology in an attempt to estimate the production of month t before the month is over; to do so we make use of daily electricity data. This is what we call estimating production in real time.
Although we employ daily data, we do not try to estimate daily production-if for no other reason than that there is no independent daily production index to provide a benchmark. We use the daily data only to get information on economic activity in a subset of month t , long enough to supply valuable information on month t as a whole but sufficiently short t o allow us to generate significantly earlier estimates. It turns out that, with a number of adjustments, it is possible to obtain reasonably accurate forecasts.
The plan of the paper is as follows. In the next section we describe the data and several different ways of using them to estimate production. In this context we also discuss how to adjust for trading-day variations. In the third section we evaluate the forecasting performance