Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/27328
標題: 台灣虱目魚及吳郭魚價格預測之研究
A Study on Price Forecasting of the Milkfish and Tilapia
作者: 黃保諭
Huang, Pau-Yu
關鍵字: Milkfish;虱目魚;Tilapia;Price Forecasting;Three-Stage Least Squares;System Method;ARIMA;Price Simulation;Aquatic Products by Breeding;吳郭魚;價格預測;三階段最小平方法;體系法;ARIMA時間數列分析;價格模擬;漁產品養殖
出版社: 農業經濟學系
摘要: 
摘 要
漁產品養殖受其生物性之影響,其生產量對價格之反應存有落遲現象,產生經濟理論之蛛網現象,此致漁產品價格大幅震盪,影響漁民所得水準及產業未來發展。因此政府對於重要之漁產品,特別制定價格穩定政策乃加以保障漁民所得及穩定民生物價。惟其前提必須先掌握價格及產量變動趨勢,才得以擬定出適當政策及正確生產計劃。故本研究主要目的為提昇漁產品價格預測能力,以保障漁民及消費者權益及健全產銷預警制度。基此,本文建立虱目魚及吳郭魚產業之供需計量模型,並配合簡單預測、遞推預測、ARIMA預測法,以體系法之三階段最小平方法,將供給、逆需求、運銷關係式作計量實證分析,探討並評估預測績效。
茲將結果歸納如下:
一、實證結果:
虱目魚及吳郭魚苗價格以及飼料價格,經體系法實證後呈現顯著及略為顯著之反應,建議政府相關單位應注意其價格水準以及變動趨勢,以避免養殖生產成本提高,致使漁民降低養殖之意願,進而影響到漁產品市場的供需穩定性。另方面,經實證後顯示,隨著國民所得增加,虱目魚及吳郭魚之零售價也呈現正向之反應,此結果表示兩種魚貨性質尚屬於正常財。
二、事前模擬:
ARIMA預測模式對於虱目魚及吳郭魚之產地批發價格之模擬,皆優於其他兩者模型。而簡單預測模型之模擬表現則略優於遞推預測模式。
三、事後預測:
1.虱目魚事後預測評估,以簡單預測模式之表現最佳,其次為遞推預測模式,但三種預測模式均能捕捉到組外樣本期間之趨勢變動,遞推預測模式略優。
2. 吳郭魚事後預測評估,遞推預測模式之表現能力為最佳,次為簡單預測模式。三模式均能準確掌握價格之趨勢變動,表現為水準之上。

ABSTRACT
The influence of quantities of aquatic products by breeding to their prices has a phenomenon of time lag that is reflected by its biological characteristics. The appearance of things comes into existence as the cobweb theory of economics. Consequently, the price of aquatic products by breeding has been seriously impacted, the income of fishing population affected, and development of fishery industry blocked.
Because of the conditions as set out above, the government has made a lot of price stabilizing policies about aquatic products not only to ensure a fair and reasonable income for fishing population in particular, but also maintain steady commodities prices in general. To establish correct policies and proper production plans, the most important premise we have is to catch well the trend of price and quantity variation. For this reason, the objectives of this study were to explore the supply-demand behaviors of aquatic products, so as to enhance the ability of price forecasting.
This study used three modes-Naive Mode, Extrapolation Mode, and ARIMA (Auto-Regression Integrated Moving-Average) Mode to estimate the equations of supply, demand, and marketing for Milkfish and Tilapia by Three Stage Least Squares (3SLS) of system method. Based on the results obtained from estimation, it is summarized as follows:
1. Parameters of fry and fishmeal of Milkfish and Tilapia displayed noticeable and slightly noticeable after proving by system methods. According to this result, the government should pay attention to the prices of fry and fishmeal and the trend of their variations. Otherwise, the costs of breeding will be increased, interest of breeding lowered, and the market equilibrium affected. Meanwhile, the parameter of retail price of Milkfish and Tilapia showed slightly noticeable, after proving a national income increased. The result, therefore, indicated that Milkfish and Tilapia were normal goods.
2. At the aspect of simulation, the result of ARIMA Forecasting Mode was superior to Naive Forecasting Mode and Extrapolation Forecasting Mode.
3. Regarding the price forecasting of Milkfish, on one hand, Naive Mode made the best performance of price forecasting. The second was Extrapolation. All of the three modes of price forecasting could capture the trend of price variations correctly. On the other hand, concerning the price forecasting of Tilapia, Extrapolation Mode had the most excellent performance of price forecasting among the three, though all of them could very correctly provide in hand the trend of price variations.
URI: http://hdl.handle.net/11455/27328
Appears in Collections:應用經濟學系

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.