跨國成長率收斂性研究-利用一般性相關效果估計方法

Estimation of growth convergence using common correlated effects estimation approaches

鄭仁傑(Jen-Chieh Cheng)

指導教授 ： 王健合

國立臺北大學/社會科學學院/經濟學系/碩士(2008年)

摘要

我們利用 Pesaran (2006) 所提出的一般性相關效果混合估計法 (Common correlated effects pooled estimation) 重新檢驗條件性收斂假說, 並且利用 Hansen (1999, 2000) 所提出的門檻模型 (Threshold model) 估計多重均衡成長模型。模型設定上是依照Islam (1995)所提出的動態追蹤資料成長模型 (Dynamic panel data growth model)。蒐集資料上, 國家數目有188個, 時間長度從1971到2003。 我們將蒐集的國家依照各別經濟特性分為三組, 分別為經濟合作發展組織組 (OECD), 非產油國家組(Non-oil), 和中間國家組 (Intermediate)。 分別在這三組使用最小平方虛擬變數估計法 (Dummy Variable Least Square Estimation) 和一般性相關效果混合估計法, 重新估計動態追蹤資料成長模型。 在非產油國家組應用門檻模型檢驗多重均衡成長模型。
研究結果顯示, 利用一般性相關效果混合估計法所估的收斂係數比使用最小平方虛擬變數估計法來得低。 在估計多重均衡成長模型上, 利用最小平方虛擬變數估計法所估計的結果顯示存在三區間的均衡, 但利用一般性相關效果混合估計法所估計的結果顯示只存在兩區間的均衡。

關鍵字

經濟成長 ； 追蹤資料模型 ； 門檻模型

並列摘要

This thesis reexamines the evidence on the conditional convergence theory using common correlated effects pooled (CCEP) estimation proposed by Pesaran(2006) and uses the threshold model proposed by Hansen (1999, 2000) to estimate the multiple equilibria growth model. We use the dynamic panel data growth model proposed by Islam (1995). The data set includes 188 countries over 1971 to 2003.
We group countries into three samples, OECD, Non-oil, and Intermediate sample, according to their economic characteristics. In these three samples, we use Dummy Variable Least Square (DVLS) and CCEP estimations to estimate the dynamic panel data growth model. In Non-oil sample, we apply the threshold model to estimate the multiple equilibria growth model. As the results, we show that the convergence coefficients estimated by DVLS are less than the coefficients estimated by CCEP. The results of estimating the multiple equilibria growth model indicate that there exists three regimes when using DVLS estimation, but there exists two regimes when using CCEP estimation.

並列關鍵字

Economic growth ； Panel data model ； Threshold model

參考文獻

1. Barro, Robert J. (1991), “Economic Growth in a Cross-Section of Countries,” Quarterly Journal of Economics, 106(2), 407–443.

2. Barro, Robert J. and Xavier Sala-i-Martin (1992), “Convergence,” The Journal of Political Economy, 100(2), 223–251.

3. Baumol, William,J. (1986), “Productivity Growth, Convergence, and Welfare: What the Long-Run Data Show,” The American Economic Review, 76(5), 407–443.

4. Breusch, T.S., and A.R., Pagan (1980), “The Lagrange Multiplier Test and its Application to Model Specifications in Economic,” Review of Economic Studies, 47(1), 1072–1085.

5. Cohen, Daniel and Marcelo Soto (2007), “Growth and human capital: good data, good results ,” Journal of Economic Growth, 22(1), 51–76.