NOISE ATAU KEDATANGAN INFORMASI: SEBUAH FENOMENA SPESIFIK PERILAKU HARGA SAHAM DI PASAR MODAL INDONESIA Studi Empiris Berbasis Data Intraday, Bursa Efek Jakarta 1999-2006

Abstract

This research proved occurrence of noise. This research used intraday data in JSX
(Jakarta Stock Exchange). Samples of the data are the firms listed in LQ 45 indexes for the year
of 1999-2006. The noise, in accordance with previous concepts and theories, were influenced by
the arrival of public and private information and those arrivals were disseminated. Test results
concluded that noise over trading and nontrading period, along with overnight and lunch break
nontrading session, and the first and second trading session, had occured.
Factually, noise occurred in the interval of one and three aforementioned periods.
Conversely, information arrival (consistently positive return) occurred in the lag of four
preceding period or one day lag only. Sequentially, this research conducted to control using size,
trading volume, bid-ask spreads, up-down market, and tick size statute. Having controlled, this
research found that these were not always correct and valid. It means that conclusions of the
prior researches were not consistent. Especially, this research suggested contra evidence in
comparisons with previous concepts and theories whenever controlled by size, trading volume,
bid-ask spreads, up-down market and tick size.
Keywords: intraday data, trading and nontrading periods, noise, negative autocorrelation, size, trading volume, bid-ask spreads, up-down market, tick size.