編輯推薦
• Active learning with highly accessible introductory text using computers and web site support rather than passive reading
• Well-prepared Excel (R) workbooks enable easy Monte Carlo simulation and other analyses
內容簡介
This highly accessible and innovative text (and accompanying website: www.wabash.edu/econometrics) uses Excel (R) workbooks powered by Visual Basic macros to teach the core concepts of econometrics without advanced mathematics. It enables students to run monte Carlo simulations in which they repeatedly sample from artificial data sets in order to understand the data generating process and sampling distribution. Coverage includes omitted variables, binary response models, basic time series, and simultaneous equations. The authors teach students how to construct their own real-world data sets drawn from the internet, which they can analyze with Excel (R) or with other econometric software.
作者簡介
Humberto Barreto is DeVore Professor of Economics at Wabash College, Indiana. He received his Ph.D. from the University of North Carolina at Chapel Hill. Professor Barreto has lectured often on teaching economics with computer-based methods, including the National Science Foundation's Chautuqua program for short courses using simulation. He has received the Indiana Sears Roebuck Teaching Award and the Wabash College McLain-McTurnan Arnold Award for Teaching Excellence. The author of The Entrepreneur in Microeconomic Theory, Professor Barreto has served as a Fulbright Scholar in the Dominican Republic. He is the manager of electronic information for the History of Economics Society and the director of the opportunities to Learn about Business program at Wabash College.
Frank M. Howland is Associate Professor of Economics at Wabash College. He earned his PhD in Economics from Stanford University. Professor Howland was a visiting researcher at FEDEA on Madrid in 1995-96. His academic research focuses on college savings plans.
精彩書評
"Barreto and Howland have taken a truly innovative approach to teach undergraduate econometrics, using computer simulation methods to illustrate and clarify difficult topics. Fully integrated with Microsoft Excel, this textbook forces students to take a hands-on approach to the subject. There is no better way to learn econometrics than by doing econometrics!"
--Jason Abrevaya, Purdue University
"Barreto and Howland have done an excellent job of producing an introductory econometric textbook based on Excel software combined with a well written and applied intuitive approach to econometrics. In my opinion, their teaching philosophy is absolutely the correct method: Put the student in front of a computer and teach econometrics by doing econometrics"
--Daniel V. Gordon, University of Calgary
"The authors wrote a textbook on introductory econometrics which is different from most textbooks by using Monte Carlo simulation with Microsoft Excel. The book is written for undergraduate students in econometrics who should not be explicitly confronted with formal mathematics but instead with visual explanations of abstract ideas."
-- Zentralblatt MATH
"Hats off to Barreto and Howland for a clearly-written text that introduces the undergraduate to data analysis and econometric techniques using Excel. The book's strength is in using Monte Carlo simulation to illustrate sampling theory and the Gauss Markov theorem. I am in total agreement with the authors that computer-based exercises help to make abstract concepts operations and meaningful. Most juniors and seniors are familiar with the basic features of Excel spreadsheets. Showing them how to use SOLVER, the DATA ANALYSIS TOOLS, and to run Monte Carlo simulations, allows an instructor to take a familiar tool (Excel) and use it to introduce undergraduates to econometrics in an intuitive and non-threatening way."
-- Jon M. Conrad, Cornell University
目錄
1. Introduction
Part I. Description:
2. Correlation
3. Pivot tables
4. Computing regression
5. Interpreting regression
6. Functional form
7. Multivariate regression
8. Dummy variables
Part II. Inference:
9. Monte Carlo simulation
10. Inferential statistics review
11. Measurement box model
12. Comparing two populations
13. The classical econometric model
14. The Gauss Markov theorem
15. Understanding the standard error
16. Hypothesis testing and confidence intervals
17. F tests
18. Omitted variable bias
19. Heteroskedasticity
20. Autocorrelation
21. The series topics
22. Dummy dependent variables
23. Bootstrap
24. Simultaneous equations.
計量經濟學導論:濛特卡洛模擬方法與應用 作者: [此處可填入原書作者信息,若無則省略] 齣版社: [此處可填入齣版社信息,若無則省略] 裝幀: 精裝 --- 內容提要 本書旨在為讀者提供一個紮實而實用的計量經濟學基礎,重點聚焦於如何運用現代統計工具,特彆是濛特卡洛(Monte Carlo)模擬方法,來解決復雜的經濟學問題和數據分析挑戰。本書不僅涵蓋瞭計量經濟學的核心理論,更強調實際操作與應用,使讀者能夠將抽象的統計概念轉化為可操作的分析模型。 本書的結構設計清晰,從基礎的概率論和統計學迴顧開始,逐步深入到迴歸分析的各個方麵。不同於傳統教材側重於證明和理論推導,本書大量引入實際經濟數據案例,並通過逐步構建和運行模擬實驗,直觀地展示瞭統計推斷背後的邏輯。我們相信,通過親手操作和觀察模擬結果,學習者能更深刻地理解統計方法的有效性和局限性。 第一部分:基礎迴顧與計量經濟學基石 第一章:計量經濟學的視角與數據類型 本章首先界定計量經濟學的核心任務——利用統計方法量化經濟理論並檢驗假設。我們將詳細區分截麵數據(Cross-Sectional)、時間序列數據(Time Series)和麵闆數據(Panel Data)的特徵及其在模型構建中的不同考量。重點討論瞭測量誤差、樣本選擇偏誤等在經濟數據中普遍存在的問題,並引入瞭描述性統計工具箱,為後續的推斷做好準備。 第二章:綫性迴歸模型(OLS)的理論基礎 本章深入探討瞭最基本的工具——普通最小二乘法(OLS)。我們詳細闡述瞭高斯-馬爾可夫(Gauss-Markov)定理的假設條件,並解釋瞭在綫性模型中,無偏性、一緻性和有效性這三個核心性質的含義。不同於純理論的論述,本章側重於解釋當這些假設被違反時(如異方差性或自相關),OLS估計量的性質如何變化,並初步引入瞭如何使用軟件進行診斷性檢驗。 第三章:OLS估計量的統計推斷 統計推斷是計量經濟學的靈魂。本章構建瞭在標準假設下檢驗參數估計量的理論框架,包括t檢驗、F檢驗的構建與解釋。我們詳細講解瞭置信區間和顯著性水平的概念,並指導讀者如何準確地解讀迴歸結果中的$p$值。此外,本章還涵蓋瞭模型設定的重要性,如函數形式的選擇(對數、平方項)對解釋力的影響。 第二部分:計量模型的擴展與挑戰 第四章:多重共綫性與模型選擇 在實際應用中,解釋變量之間的高度相關性(多重共綫性)是常見難題。本章深入分析瞭多重共綫性的後果——估計量的方差膨脹,而非估計量本身的偏誤。我們將探討診斷多重共綫性的方法,例如方差膨脹因子(VIF),並討論在麵臨共綫性問題時,模型簡化或變量選擇的策略。 第五章:異方差性(Heteroskedasticity)的處理 異方差性是指誤差項的方差不恒定。本章係統闡述瞭異方差性對OLS估計量的影響——估計量仍然無偏且一緻,但標準誤的計算是錯誤的,導緻統計推斷失效。我們將重點介紹處理異方差性的方法,包括加權最小二乘法(WLS)和使用穩健標準誤(Robust Standard Errors),並提供在不同情境下選擇閤適方法的指南。 第六章:時間序列數據的初步探討 本部分將視角轉嚮時間序列數據。我們將介紹時間序列數據的基本特徵,如平穩性(Stationarity)的概念及其重要性。本章構建瞭自迴歸(AR)和移動平均(MA)模型的初步框架,並講解瞭如何通過平穩性檢驗(如ADF檢驗)來識彆時間序列的內在結構。 第三部分:濛特卡洛模擬法的核心應用 第七章:濛特卡洛模擬導論 本章是全書的重點之一,係統介紹濛特卡洛(MC)模擬方法的原理。我們將從隨機抽樣、僞隨機數生成器的特性講起,解釋如何通過重復模擬來逼近真實的概率分布和期望值。本章將詳細展示如何利用電子錶格軟件(如Microsoft Excel)強大的矩陣運算和隨機數生成功能,搭建初步的模擬環境,例如模擬拋硬幣的次數分布。 第八章:利用濛特卡洛模擬進行參數估計和檢驗 我們將把MC方法應用於計量經濟學的核心問題。首先,演示如何利用MC模擬來驗證在有限樣本下,OLS估計量的漸近性質是否成立。隨後,重點講解如何通過模擬重采樣(Bootstraping)技術,來計算在存在復雜異方差或非正態誤差項時,參數估計量的更準確的標準誤和置信區間。這使得讀者可以超越依賴嚴格理論假設的局限。 第九章:處理復雜模型的穩健性分析 在更復雜的模型結構中,解析解往往難以求得。本章展示瞭MC模擬在處理這些“棘手”情況下的威力。例如,我們將模擬在非綫性約束或存在高階滯後變量的模型中,估計參數的分布情況。通過大量模擬,我們可以評估不同估計策略的穩健性,並對模型設定進行敏感性分析,瞭解關鍵參數對模型假設變化的反應程度。 第十章:政策評估與風險分析的模擬 本章將計量經濟學與實際決策相結閤。我們將構建一個簡單的宏觀經濟模型,通過設定不同的政策衝擊(如財政支齣增加),並利用濛特卡洛模擬來評估這些政策對關鍵經濟指標(如通貨膨脹、失業率)的長期影響分布。這為讀者提供瞭一種量化政策不確定性和風險暴露的實用工具。 結語:邁嚮高級計量經濟學 本書的目的是為讀者打下堅實的理論基礎,並武裝以強大的數值模擬能力。掌握瞭濛特卡洛方法,讀者便能更自信地處理現實世界數據所固有的復雜性和不確定性。本書為後續學習因果推斷、麵闆數據模型或更高級的時間序列分析,如嚮量自迴歸(VAR)模型,奠定瞭不可或缺的實踐基礎。 --- 本書特點: 理論與實踐並重: 既有嚴格的統計學理論支撐,又緊密結閤實際經濟案例。 操作性強: 詳細指導讀者如何利用普及性軟件(Microsoft Excel)實現復雜的模擬過程。 側重現代方法: 重點介紹對當代經濟分析至關重要的濛特卡洛與重采樣技術。 案例驅動學習: 所有關鍵概念都通過具體的經濟數據實例進行演示和驗證。