Change search
ReferencesLink to record
Permanent link

Direct link
Investigating the Accuracy of Analyst Consensus for Earnings per Share of S&P 100 companies
KTH, School of Engineering Sciences (SCI), Mathematics (Dept.), Mathematical Statistics.
KTH, School of Engineering Sciences (SCI), Mathematics (Dept.), Mathematical Statistics.
2015 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

This study investigated what affects how accurately financial analysts can predict the earnings per share of companies included in the Standard & Poor’s 100 index. To achieve this goal data on earnings forecasts was gathered for the years 2000 through 2013. Further this study investigated if there are any differences in the accuracy of optimistic respectively pessimistic earnings consensus forecasts. Multiple linear regressions were used in order to answer the imposed questions. The factors found, in this study, to affect the forecast accuracy were consensus dispersion, company size, net margin, which exchange a company is listed on and, to some extent, a company’s industry classification. Further, the result of the study implies that there is no difference in the accuracy of optimistic respectively pessimistic earnings consensus forecasts. In this study the models utilised, and the factors investigated, could only explain a limited part of what affects the earnings forecast accuracy. Lastly, a concluding qualitative attempt was made to find factors which affect the accuracy but which were hard to incorporate in this quantitative study.

Abstract [sv]

Denna studie syftade till att undersöka vad som påverkar hur noggrant finansanalytiker kan förutsäga resultatet per aktie för företag som ingår i Standard & Poors 100 index. För att uppnå detta mål samlades data om vinstprognoser in för åren 2000 till 2013. Vidare syftade denna studie till att undersöka om det finns någon skillnad i noggrannhet hos optimistiska respektive pessimistiska konsensusprognoser. Multipla linjära regressioner användes för att besvara dessa frågor. De faktorer som denna studie fann har påverkan på prognosprecisionen är konsensusdispersion, företagsstorlek, nettomarginal, vilken börs företaget är noterat på och, i viss mån, ett företags industriklassificering. Vidare visar resultaten av studien att det inte finns någon skillnad i noggrannheten för optimistiska respektive pessimistiska konsensusprognoser. I denna studie kunde modellerna som användes, och faktorerna som undersöktes, bara förklara en begränsad del av vad som påverkar en resultatprognos noggrannhet. Slutligen fördes en kvalitativ diskussion kring vilka faktorer som påverkar träffsäkerheten ytterligare, men som var svåra att implementera i denna kvantitativa studie.

Place, publisher, year, edition, pages
2015. , 52 p.
National Category
URN: urn:nbn:se:kth:diva-168017OAI: diva2:813895
Available from: 2015-05-25 Created: 2015-05-25 Last updated: 2015-05-25Bibliographically approved

Open Access in DiVA

fulltext(1205 kB)102 downloads
File information
File name FULLTEXT01.pdfFile size 1205 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
Mathematical Statistics

Search outside of DiVA

GoogleGoogle Scholar
Total: 102 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 432 hits
ReferencesLink to record
Permanent link

Direct link