eThesis - School of Business electronic theses

The purpose of this thesis is to study the differences in forecasting process of stock analysts as the industry of the target company changes and the scrutiny includes different financials. Therefore, this paper studies particularly the forecasting processes in Software & IT Services and Industrial Commercial Services industries and compares them with a sample of general market constituents.

The study exploits historical forecasting data obtained from I/B/E/S analyst database and actual reported financials obtained from Compustat database. The comparison includes companies that had continuous 12 month forecast for all the three financial metrics included from the beginning of 2005. These key financials are Net Sales, EBITDA and EPS. The division of companies into different industries relies on Thomson Reuters industry classifications.

This paper finds the analyst forecasting process to differ among different industries when compared against each other and a sample of the general market constituents. The greatest difference in forecasting process of Software & IT Services industry was found in EBITDA forecasting where the stock analysts place significant weight on the previous annually reported figure. As for the Industrial Commercial Services industry it could be observed that stock analysts seem to emphasize previous reported figures in Net Sales forecasts.

Finally EPS forecasting records a differing forecasting process than the other two metrics as it indicates an emphasis to be laid in the median analyst estimate from previous month with the greatest weight placed on the Industrial Commercial Services industry's prior forecasts. In similar manner, the descriptive study indicates this industry to contain the least forecasting errors within time among the groups incorporated in this study. Furthermore, Software & IT Services industry seems to carry the greatest positive forecasting error as expected.