糧食是維持人類生存的基本資源，而工業化社會亦脫離不了對石油的需求。因 此，當糧食或石油的供需出現不穩定狀況，將對一國甚或全球經濟造成衝擊。此外， 隨著國際金融市場越趨開放，匯率雖可能受一國政府左右，但匯率的升眨亦使一國面臨物價穩定與經濟成長之間的抉擇。過去文獻討論經濟成長議題時，大多以單一因素 對經濟成長的衝擊，而本研究以油價、糧價作為實質原物料價格，再加上實質匯率因 素對經濟成長的影響，以期能獲得較為周延的結果。本文利用橫斷面的異質差異並加 入多變量的架構，採用（Pesaran，Shin, & Smith, 1999)的混合組群平均估計式（PMG) 進行分析，除檢測是否存在長期均衡關係，並探討是否因為長、短期的影響而有所差 異？並與(Pesaran & Smith, 1995 )提出的組群平均（MG)及傳統的動態固定效果（DFE) 估計方法比較，再經由Hausman檢定選擇較合適的估計方法。實證結果顯示，採用 PMG的估計方法較為有效，實質原物料價格及實質匯率與經濟成長存在長期關係，且 長、短期的影響效果不同。因此，探討經濟成長時，應同時考慮實質原物料價格及實 質匯率等因素，以避免產生錯誤的結論。另外，分別就高收入與低收入不同經濟結構 體進行分析，發現實質原物料價格與實質匯率對經濟成長亦有不同的影響效果。

The economic growth is considerably influenced by energy and food crises since food and oil are the fundamental resources for humans. In addition, the exchange rate is also a crucial factor for economic growth. Different from the previous reports that focused mainly on the impact of a single factor on economic growth rates, this study emphasizes a multiple factor analysis including the real oil prices, food prices and exchange rates. The purpose of this paper is to investigate the long- and short-run impact of three factors, i.e. oil prices, food prices, and exchange rates on economic growth rates. To this end, we employed the pooled mean group estimator (PMG) from Pesaran, Shin, and Smith, 1999, the mean group estimator (MG) from Pesaran and Smith (1995), and the dynamic fixed-effect method (DFE) to balanced panel data for 53 countries over the period 1996-2011. In addition, the Hausman test was used to choose the more suitable estimator for this analysis. In the empirical results, the PMG estimator is shown to be more suitable for this analysis. Moreover, the negative long-run relations between these factors and economic growth rates coexist with positive short-run relations. Notably, we found that the lack of simultaneous consideration of these factors results in a substantial deviation in the analysis. Another finding is that the results are consequently varied when different income groups are included in the data analysis. Therefore, a consideration of the multiple factors including the real oil prices, food prices, exchange rates, as well as the income groups, is needed for an unbiased impact analysis of economic growth rates.