STRUKTUR CORPORATE GOVERNANCE DAN KETEPATAN WAKTU PENYAMPAIAN LAPORAN KEUANGAN: PERBANDINGAN MODEL LOGISTIK DAN NEURAL NETWORK

Abstract

The main objective of this study was to examine the impact of corporate gover-nance structure and the performance of the firms to timelines in Indonesian Stock Exchangeusing two alternative methods, Logistic Regression and Neural Network. This study com-bined corporate governance structure and timelines study. Samples in this study were publiccompanies listed in Indonesian Stock Exchange. The dependent variable was timelines proxiedby dummy variable, 1 if companies published financial reporting before 120 days after De-cember 31 and 0 otherwise. Governance structures are proxied number of the Board ofDirectors and number of the Board of Commissioners. The results of the study showed thatthe prediction accuracy of logistic regression is 61.2% while Neural Network is more than96%. This suggested that Neural Network predicts more accurately than logistic regression.