| Summary: | 碩士 === 中原大學 === 國際貿易研究所 === 98 === Recent literatures started to focus on multiproduct firms and explore the effect of product-level determinants on firm behavior. Most of the extant empirical researches about product switching use census data and binary choice method. However, census data are generated every five years. Using five-year data is easy to loss valuable information relative to using the yearly data. In addition, it is not able to describe clearly about product switching of multiproduct firms when the binary model is utilized. In view of this, this paper tries to use product allocation and yearly data to improve existing questions.
There are two different ways between this paper and past literatures. The first is that we use yearly data to replace census data. Yearly data possess more detailed information than census data. Second, we use the continuous product share to replace the binary choice. Compare to the discrete binary choice, product share reveals the continuity of product choice.
The studies on the product choice are interested in the effect of product similarity, age as well as firm scale. This paper extends to investigate the impact of these factors on product share. Our data from textile publicly traded company in Taiwan over the period of 1992 to 2008. The empirical results show the product similarity, age, firm scale, market competitions significantly affect product share, but business cycle is not a significant factor for product share.
|
|---|
