 
|
|
|
|
| Author |
Carlsson, Sven, 1981-
|
| Title |
Are Security Analyst Forecasts Bulletproof? : A Study of Security Analysts Accuracy in Forecasting Quarterly Earnings / Sven Carlsson, Anders Hillerborg |
| Published |
Stockholm Handelshögskolan i Stockholm 2008 |
|
|
|
|
|
|
|
| Subjects |
Master Thesis in Accounting and Financial Management
|
| Subjects |
Security Analyst Forecast
|
|
Univariate Time-Series Models
|
|
Earnings Forecast
|
|
Forecasting Accuracy
|
|
Earnings Forecasting Accuracy
|
| Pages/vol |
67 s. |
| Author/editor |
Hillerborg, Anders, 1982-
|
| Organisation |
Handelshögskolan i Stockholm Institutionen för redovisning
|
| Summary |
The purpose of this Master Thesis is to evaluate the contribution of Security Analysts earnings forecasts, by comparing their forecasting accuracy with that of Time-Series Models, and by test if and how their forecasting accuracy varies with different circumstances. We have studied Security Analysts forecasts on quarterly earnings data of Swedish companies between 2001 and 2007. We have measured the forecasting accuracy in two ways; (1) using Percentage Error Analysis, which has been the most commonly used evaluation tool in previous research, and (2) Regression Line Analysis, which is used in order to reduce the measurement bias caused by the size of earnings when using the Percentage Error approach. We find that Security Analysts are more accurate than Time-Series Models when forecasting quarterly earnings, although they tend to underestimate earnings. The forecasting accuracy of Security Analysts varies with different circumstances. We conclude that the accuracy varies between different sectors, and that the most accurate forecasts are those made for companies in the healthcare sector. For the remaining circumstances investigated in this thesis, namely the size of the company, the forecasted quarter and the market trend, we do not find any statistically significant differences in Security Analysts forecasting accuracy. |
|
|
|
|
|
|