本研究探討2000年1月至2011年12月間台灣股票市場中,營收創新高家數與總體變數(台灣貨幣供給額(M1B)、消費者物價指數、工業生產指數、失業率、台灣加權股價指數、倫敦黃金現貨價、西德州原油現貨價及金融業隔夜拆款利率)之關連性,其中,營收創新高家數包含台灣所有上市、上櫃的營收創新高家數,該變數依據資料庫中每月的營收創新高家數作加總。 本文採用Zhu(2011) A negative binomial integer-valued GARCH model(負二項整數GARCH模型;簡稱NBINGARCH模型)來觀察變數間之變化,並使用該模型預測下個月營收創新高家數,另透過Theil’s U值分析NBINGARCH模型的預測能力。實證結果:(1)當期台灣貨幣供給額(M1B)、落後一期消費者物價指數、當期工業生產指數、落後一期失業率、當期的台灣加權股價指數、落後一期的西德州原油現貨價及當期的金融業隔夜拆款利率與營收創新高家數呈顯著的正向關係;(2)落後一期台灣貨幣供給額(M1B)、當期的失業率、落後一期台灣加權股價指數及當期倫敦黃金現貨價呈顯著的負向關係。(3)實證顯示NBINGARCH模型預測能力具參考價值。
The purpose of this paper is to explore the relationship between number of high revenue innovation and macroeconomic variables (M1B, Consumer Price Index, Industrial Production, Labor Force Participation Rate, Unemployment rate, Taiwan stock index, gold price, oil price and interest rate) in Taiwan stock market from 2000 to 2011. The number of high revenue innovation include Taiwan Stock Exchange Capitalization Weighted Stock Index, the variables are sum of each month that the number of high revenue innovation in data. This paper uses Zhu(2011) A negative binomial integer-valued GARCH model to observe change of variables. Use the model to predict the next month of number of high revenue innovation. Through the Theil's U value analysis prediction capability of NBINGARCH model. The empirical evidence demonstrates as follows: 1. Positive correlation appears between current Monetary Aggregate M1B, lagged Consumer Price Index, current Industrial Production Index, lagged Unemployment Rate, current TAIEX, lagged West Taxes crude oil spot price, current Interbank Call Overnight Interest Rates and the number of high revenue innovation. 2. Negative correlation appears between lagged Monetary Aggregate M1B, current Unemployment Rate, lagged TAIEX, current Loco London gold spot price and the number of high revenue innovation. 3. The evidence demonstrates the NBINGARCH model of predictive ability has reference value.