Description

With real-time data it is analyzed what information Chile’s monthly indicator of economic activity (IMACEC) contains about the final GDP, defined as the growth rate that has been subject to at least two annual revisions. Data are presented and revisions briefly analyzed. It is argued that when three months of IMACEC data are available, it is possible to extract signals about the final GDP, which are as reliable as those contained in the first release of the growth rate. This result is obtained with the evaluation in-sample as well as out-of-sample. It is then investigated how much extra information IMACEC data provide of the final GDP compared to what is already present in historical data. The in-sample analysis indicates statistically significant improvements when more IMACEC data of the quarter are available. Measured by the root mean square nowcast error (RMSNE) the out-of-sample performance also improves as more monthly data are published, although when only the first IMACEC data of the quarter are available, this is not statistically significant.