An empirical study on the day-of-the-week effect of financing-oriented commercial papers in Taiwan’s money market

• Main Authors: Chung-Wei Chao, 趙仲偉
• Other Authors: Jack J.W.Yang
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/12977726070963302479

碩士 === 國立雲林科技大學 === 財務金融系碩士班 === 91 === Taiwan promotion of economic growth while facing finance management notions changed greatly over the past decade. In behalf of manufacturers are deeply affected as they’re founding through the fluctuating level of interest rates. This paper probes into the market abnormal returns of the day-of-the-week effect in terms of comprehending the market objects as decision-making in financing policy. The basis of this paper is to verify the financing commercial papers day-of-the-week effect phenomenon. To do this, we used Ordinary Least Square (OLS) analysis, autoregressive conditional heteroskedasticity (ARCH) modeling strategy, introduced by Engle (1982), and general autoregressive conditional heteroskedasticity (GARCH) models to study three cataloged data divided into three periods (one-day-off per week, two-day-off twice a month, and two-day-off per week) from Taiwan Economic Journal Data Bank. 1. Both of the three models shown 30-day period Thursday effect and 180-day period Monday effect in one-day-off per week of weekday effect. 2. There are no identical effects as featured by the three models in two-day-off twice a month. 3. A Monday-effect emerged from 10-day, 20-day, 90-day, and 180-day period by three models in two-day-off per week model. In the point of dispatching of founding for manufacturer, the best choice is to select Monday to issue commercial paper in terms of achieving the purpose of lowering interest rate. 4. A construction variation appeared to the week effect of commercial paper after enforced the two-day-off per month. Consequently, this study proven that financing commercial papers exist in the day-of-the-week effect, not conform to weak form efficient market hypothesis.