本文旨在量化廣告支出與總銷售額的線性關係，應用縱橫資料模型分析台灣產業從1993年到2013年之廣告銷售彈性。本文使用的樣本包含27種台灣各個不同的產業，分別估計各產業的廣告銷售彈性。研究結果顯示，在Pooled OLS模型估計之下的廣告銷售彈性為0.28，而使用縱橫資料的固定效果模型估計廣告銷售彈性得到的結果為0.26，最後改用動態的縱橫資料模型並以兩階段估計法做估計，得到短期的廣告銷售彈性為0.07，而長期的廣告銷售彈性為0.13。為觀察總體變數對銷售額的影響，加入總體變數進行估計，研究結果顯示，實質所得與銷售額呈現正向相關，實質產業價格與銷售額呈現負向相關。27個產業中，廣告支出占總銷售額的比率最高之兩產業分別為食品業及生物科技與醫療產業，比較其廣告銷售彈性以生技醫療產業為大。 The goal of this thesis is to explore quantification of a linear statistical relation between advertising and total sales. We apply pooled OLS regression and panel data analysis including fixed effects regression and dynamic panel data regression to estimate the advertising elasticity of sales for Taiwan industries from 1993 to 2013. This thesis contains 27 different industry and estimate the advertising elasticity of industry individually. We find the advertising elasticity is 0.28 by the pooled OLS model and the fixed effects panel model produces an estimate of 0.26. To estimate by dynamic panel data model, the short-term advertising elasticity is 0.07 and the long-term advertising elasticity is 0.13 using a two-step estimation method. From examination of the relative factors, we find the real income has a positive correlation with sales and the real industry price has a negative correlation with sales. Among the 27 industries, the highest industry of advertising expenditure to sales ratios is the bio-technology and medical care and food industries. With respect to the advertising elasticities of sales for these two industries, the bio-technology and medical care industry has a higher estimated elasticity than that of the food industry.