A multi-industry analysis of structural changes and earnings forecasts

The purpose of this study was to examine whether structural change modeling procedures improved the predictive ability of quarterly earnings-per-share across industries. The structural change detection methods derived by Chen (1984), Chang, Tiao, and Chen (1988), Chen and Tiao (1990) and Lee and Chen (1990) were employed in order to determine the time points when structural change occurred. These interventions were then included as intervention transfer functions in ARIMA models.