This thesis observes and discusses the link between the dispersion in analysts’ earnings forecasts (AEFS) and accounting quality (AQ). In our research we have used literature that makes the results useable in several further contexts even though it is concentrated on Danish data. Compared with previous studies we adopt not only earnings quality as the substitute variable for accounting quality but also alternative measurements for earnings quality, the dispersion of analysts forecast and different models to detect discretionary accruals. By exploring a marked based depended variable (AEFS) by an accounting based independent variable (AQ) we seek a highly plausible correlation result by the use of both Spearmans - and Pearsons Correlation Matric. Surprisingly we find that the dispersion in analysts’ earnings forecasts is positively correlated to accounting quality, which fills in the blank on the forecasting literature in Denmark. This conclusion holds significantly for the Standard Jones Model, the Modified Jones and Dechows Model in the period 2001-2010. Furthermore several robustness checks support these findings. Our findings contribute not only to the literature by illustrating the impact IFRS has on the amount of earnings management. It is also questioning the practical grip of one of the basic intuitions in the forecasting literature such as the implied negative relationship between AEFS and AQ. The research suffers from the low amounts of observations available in Denmark, which affect the possibility of getting significant results. It would therefore be interesting to expand it towards an international perspective. We hope that the findings in this paper will protect the individual investor by enhancing the stock market efficiency and further, that it will create another fundament for further research on the inputas well as the output side in understanding the incentives of analysts’ forecasts’ and their decision process.
|Educations||MSc in Finance and Accounting, (Graduate Programme) Final Thesis|
|Number of pages||137|