1.

SMAILOVIĆ, Jasmina, GRČAR, Miha, LAVRAČ, Nada, ŽNIDARŠIČ, Martin (2016). Dynamic parameter adaptation of SVM based active learning methodology. V: KREMPL, Georg (ur.). AL@iKNOW 2016 : proceedings of the Workshop on Active Learning: Applications, Foundations and Emerging Trends co-located with International Conference on Knowledge Technologies and Data-driven Business (i-KNOW 2016), Graz Austria, October 18, 2016, (CEUR workshop proceedings, vol. 1707). [S. l.]: CEUR-WS, 2016.

Dynamic parameter adaptation of SVM based active learning methodology

COBISS.SI-ID: 29892903

2.

Podatkovna baza 37GB (korpus 4 mio sporočil Twitter, korpus letnih poročil velikih, na borzi kotirajočih podjetij za zadnjih 10 let)

Baza podatkov za raziskave.

3.

Valuation and forecasting roles of dividends of Indonesian listed firms

We study the role of dividends in valuation and in forecasting future earnings in a low-protection environment with highly concentrated ownership that is expected to yield low earnings quality. Using a sample of 372 distinct Indonesian firms listed on the IDX during the period 1995 to 2012 we show that dividends are reliably positively priced by the capital market, violating thus the dividend displacement theorem. This result persists even after controlling for some typical factors that affect firm value (capital structure, risk) and the effect of factors not separately identified, but priced by the Indonesian capital market (other information). Dividends replace accounting earnings entirely in valuation. Dividends are positively correlated to future earnings over and above current accounting earnings and other accounting and market variables. Both findings show that dividends play a central role on the Indonesian capital market. Finally, we show evidence consistent with the view that earnings management of Indonesian firms after 2002 is contractually efficient rather than opportunistic.

COBISS.SI-ID: 22871270