臺灣博碩士論文加值系統

由於過去傳統農業的產業分析與預測模型主要專注於單一農產，除了農委會依產值計算的農業指數之外，沒有其他綜合指數，因此此論文將參考其他非農業之綜合指數的計算方式建構農業的綜合指數，並希望該指數能為農企業與政府提供更全面與精準的投資參考依據。透過文獻回顧，本論文將以試編農業採購經理人指數為目標，並驗證結果與總產值的相關性，藉此了解該指數的準確性。
藉由研究台灣採購經理人指數的編制方式與報告，本論文將參考其作法，並以2016年台灣產值最高的產業—養豬產業做為範例，使用農委會與中華食物網等數據試編養豬產業的採購經理人指數。
針對養豬產業篩選APMI的變數時，本論文使用線性回歸與主成分分析檢驗變數之間的關聯性與對整體變異的貢獻度，並以此貢獻度作為APMI模型的權重。
本論文總共跑了12次PCA，前六次以總交易頭數、每頭重量、每公斤價格、屠宰頭數、總重量、養豬戶數，六個變數做篩選；後六次增加四個新的變數，分別為:玉米價格、黃豆粉價格、飼料成本比、豬糧比。主成分分析後，本論文使用主成分分析的12種不同的變數與權種組合算出APMI，其中，第四組APMI使用了四個變數，分別為: 總交易頭數、每公斤交易價格、總重量與養出戶數。與養豬產業總產值的相關性高達0.7028，此組變數為目前算出來的APMI中最顯著與養豬產業總產值相關的一組。

本論文的預期貢獻有三: 1. 透過文獻回顧與實際驗證，尋找適合試編APMI的方式 2. 透過訪問養豬產業專業人士，選出適合養豬產業的APMI變數 3. 驗證養豬產業的APMI與總產值的相關性，並未未來的相關研究提供參考數據。

Most of the agricultural researches and forecasting in Taiwan are focusing on individual agriculture industry or product. Hence, there is an absence of agricultural index to describe the economic status in the market. Agricultural companies and government departments haven’t had information except agricultural Gross Domestic Profit (GDP) to measure the development of the current industry. To solve the problem, we search for index who shares the features mentioned above in non-agriculture industry, and thus, we found Purchasing Managers’ Index (PMI) and Non-Manufacturing Index (NMI).
In this thesis, we consider that PMI and NMI share the features, which we are looking for in agriculture index, are able to predict GDP variability also represent economic status of the industry. Therefore, we will conduct a trial agricultural PMI as aim of the research.
Including twelves variables monthly data for five years, we will adapt correlation coefficient to determine if the variables are strongly correlated to macro-economic indexes. After correlation coefficient, Principle components analysis will be applied to decide the best weight for each variables in APPMI (Agricultural Pig Purchasing Managers’ Index) formula.
With all diffusion indexes multiply the optimal weight for each variable and adding them up, we have APPMI as time series for five years. In twelves different combinations of data set, which had been conducted principle component analysis, we had found one of the PCA4 data set is able to interpret 70 percent of variability of total revenue in hog industry. In PCA4, variables, traded amount、price per kilogram、total weight、farms amount are included.
After refining and seasonal adjusting, also combining with other agriculture industries, APMI will be able to represent as agriculture economic status, which could be a solid reference for agricultures companies and government while making decision or evaluation for investing in the industry. In the thesis, assumptions and limits also are illustrated for further researches and refine.

摘要 i
Abstract iii
Contents iv
List of Figures vii
List of Tables vii
List of Equations ix
List of Abbreviations x
1. Introduction 1
1.1. Background 1
1.2. Research purpose 2
1.3. Research process 3
2. Literature Review 5
2.1. Background of PMI and NMI 5
2.1.1. Institution for Supply Management 5
2.1.2. Chung-Hua institution for economic research 6
2.1.3. Indicator qualities of PMI, NMI 6
2.2. PMI construction 8
2.2.1. Performance of diffusion index forecasting macroeconomic time series 8
2.2.2. ISM PMI Construction 8
2.2.3. Methodology of sampling 9
2.2.4. Weight adjusting of PMI diffusion indexes 10
2.3. PMI macroeconomic indexes and forecasting validation 10
2.4. PMI Application 10
3. Methodology 12
3.1 Process 12
3.2 How to analysis the results 18
3.2.1 Correlation coefficient and Coefficient of determination 18
3.2.2 Principle components analysis 18
3.2.3 Diffusion Index 19
3.2.4 APMI 19
4. Case study of hog industry 20
4.1 Interview with specialist 20
5. Results and Analysis 22
5.1 Assumptions 25
5.2 Results 26
5.2.1 Correlation Coefficient and Coefficient of Determination 26
5.2.2 Principle Components Analysis 29
5.2.3 Diffusion Index 35
5.2.4 Correlation of APMI and total revenue of hog industry 37
5.3 Results and Analysis 41
5.3.1 Variables Features 41
5.3.2 Principles components analysis 48
5.3.3 Correlation coefficient of APPMI and hog industry total revenue 53
6. Conclusion 55
6.1 Recommendations 55
6.2 Contributions 56
6.3 Limits 56
References 57
Appendix 59
Appendix 1: Diffusion Index and APPMI 59
The first set of variables 59
The second set of variables 68
The third set of variables 75
The fourth set of variables 82
The fifth set of variables 87
Variables and Correlation Coefficient 95
Conclusion of PCA, second round of variables test 95
PCA Second round of variables test 96
The sixth set of variables 98
The seventh set of variables 101
The eighth set of variables 103
The ninth set of variables 109
The tenth set of variables 116
The eleventh set of variables 117
The twelveth set of variables 119

Baum, Kurov, Wolfe. (2015). What do Chinese macro announcements tell us. Journal of International Money, 100–122.
Afshar, Arabian, Zomorrodian. (2007). Stock Return, Consumer Confidence, Purchasing Manager’s Index And Economic Fluctuations. Journal of Business & Economics Research, 97-106.
Cahill. (2017). Report on Business Analyst. ISM, ROB/Research.
Cho, Ogwang. (2006). A Conceptual Framework for Computing U.S. Non-manufacturing PMI Indexes. The Journal of Supply Chain, 44-53.
Cho, Ogwang. (2007). Conceptual Perspectives on Selecting the Principal Variables in the Purchasing Managers’ Index. The Journal of Supply Chain, 43-53.
Harris. (1991). Tracking the Economy with the Purchasing Managers' Index. Federal Reserve Bank of New York, 61-69.
Harris, Owens, Sarte. (2004). Using Manufacturing. Federal Reserve Bank of Richmond Economic Quarterly.
Heij, Knapp. (2014). Dynamics in the Dry Bulk Market. Journal of Transport Economics and Policy, 499–514.
Kauffman. (1999). Indicator Qualities of NAPM Report On Business. Instituation of Supply Management.
Koenig. (2002). Using the Purchasing Managers ’ Index to Assess the Economy’s Strength and the Likely Direction of Monetary Policy. Federal Reserve Bank of Dallas Economic and Financial Policy Review.
Newbold. (1974). SPURIOUS REGRESSIONS IN ECONOMETRICS. Journal of Econometrics, 111-120.
Pavur, L. a. (2005). As the PMI Turns: A Tool for Supply Chain Managers. Journal of Supply Management , 30-39.
Pelaez. (2003). A reassessment of the Purchasing Managers' Index. Business Economics.
Tsuchiya, Y. (2012). Is the Purchasing Managers' Index useful for assessing the economy's strength? A directional analysis. Economics Bulletin,. Economics Bulletin, 1302-1311.
Tsuchiya, Y. (2013 ). Purchasing and supply managers provide early clues on the direction of the US economy: An application of a new market-timing test. International Review of Economics and Finance, Forthcoming .
Tsuchiya, Y. (2013). Do corporate executives have accurate predictions for the economy? A directional analysis. Economic Modelling, 167–174.
Watson. (1998). Diffusion Index. SSRN.
Watson. (2002). Forecasting Using Principal Components From a Large Number of Predictors. Journal of the American Statistical Association, 1127-1179.
Watson. (2002). Macroeconomic Forecasting Using Diffusion Indexes. Journal of Business & Economic Statisitcs, 147-162.
Wu. (2015). Markov regimes switching with monetary fundamental-based exchange rate. Asia Pacific Management Review, 79-89.
Zhang, Chen, Wang. (2010,). A Forecast Model of Agricultural and Live-stock Products Price. Advances in Information and Communication Technology, 40-48.
陳馨蕙. (2011). The Compilation of Purchasing Managers’ Index for Taiwan. 台北: 中華經濟研究院.
陳馨蕙. (2011). 台灣製造業採購經理人指數之編制與剖析. 台北: 經建專論.