Year of manufacture: 1998

Genre: Econometrics, statistics

Publisher: Unity

Format: DjVu

Quality: Scanned pages

Number of pages: 1000

Description: The textbook covers all issues of probabilistic-statistical modeling and data analysis in economics - from elementary courses probability theory and mathematical statistics to advanced methods of multivariate statistics, time series analysis and econometrics itself. The combination and interconnected presentation of all these basic econometric disciplines in one textbook make it unique in its own way not only in the domestic, but also in the world. educational literature of this profile, allow you to structure the educational process in such a way as to achieve a holistic systemic perception of the entire block of these disciplines.

The proposed textbook reflects an understanding of the content of the mathematical and statistical tools of econometrics, which is somewhat different from the generally accepted one. In our opinion, modern achievements mathematical and statistical science (especially in multivariate statistical analysis), on the one hand, and a significant expansion of the circle economic tasks, requiring econometric methods of solution, on the other hand, necessitated a broader view of the mathematical and statistical tools of econometrics and, in particular, the inclusion in it, in addition to the traditional sections on regression models, time series analysis and systems of simultaneous equations, such sections of multivariate statistical analysis , How Markov chains, classification of multidimensional observations and reduction of the dimension of the analyzed factor space. Speaking about a wide range of economic problems that require solutions that go beyond the traditional framework of econometric methods, we meant, in particular, the statistical study of the dynamics structural changes(in demography, in the stratification structure of society, etc.), identifying hidden (latent) factors that determine the course of a particular socio-economic process, constructing integral indicators of the quality or efficiency of the functioning of the socio-economic system, typology of socio-economic objects and etc.
Secondly, during many years of experience in teaching various disciplines of probabilistic and statistical profile in economic universities and in the economics faculties of universities, we have come to the conviction that it is necessary to structure the educational process in such a way as to achieve a holistic, systemic perception of the entire block of these disciplines. It's about, in particular, about courses on elementary methods statistical processing data science (or descriptive statistics), probability theory, mathematical statistics, multivariate statistical analysis (or multivariate statistical methods), time series analysis and, finally, econometrics. Obviously, a textbook that simultaneously contains an interconnected presentation of all these courses should contribute to the realization of this goal.
In other words, we tried to write the kind of book that we would want to have at hand during our teaching activities. Unfortunately, among the many excellent foreign books on econometrics, there was no book that has the two above features.
Note that despite the presence of a number of illustrative examples and exercises, the proposed textbook does not solve the problem of the econometrics problem book. Therefore, to carry out a full educational process it should be supplemented by a set of econometric problems and exercises (for example, in the spirit of the book).
The textbook material and responsibility are distributed among the authors as follows. V. S. Mkhitaryan took part in writing chapters 6, 7, 8 and 13, and also suggested most of problems included in the chapters of the textbook. The rest of the material (including the chapters mentioned) was written by S.A. Ayvazyan. He also carried out general scientific editing of the textbook.

We continue to engage in quality management and related areas.
This time it is offered reference book on applied statistics in 3 volumes.

The author of this magnificent publication, Professor Sergei Artemyevich Ayvazyan, together with A.I. Orlov, for the first time in the USSR introduced the concept of “applied statistics,” which caused a real storm of indignation among party bosses and the top of the State Statistics Committee: statistics has always been a political matter. During perestroika, the controversy spilled onto the pages of specialty journals.

Volume 1. Fundamentals of modeling and primary data processing

The book is devoted to methods of preliminary statistical data analysis and model building real phenomenon characterized by these data. Information on probability theory and mathematical statistics is provided, and issues of software implementation are covered.
the methods presented.

Volume 2: Dependency Research

The book discusses the methods of correlation, regression and analysis of variance. Their algorithms and an overview of the software are given.

Volume 3. Classification and dimensionality reduction

The problems of object classification and dimension reduction are considered. Big
attention is paid to exploratory statistical analysis.

NATA: Books are premium, no backup needed

Topic tags:
Statistics

Publisher: Finance and Statistics

Year of publication: 1983

Pages: 472

Language: Russian

Quality: good

Using statistical methods of analysis in assessment economic activity enterprise coursework

INTRODUCTION 3

CHAPTER I. STATISTICAL ANALYSIS OF THE EFFICIENCY OF USE OF THE BASIC RESOURCES OF THE ENTERPRISE

1.1. CALCULATION AND EVALUATION OF FIXED ASSETS FLOW INDICATORS 6

1.2. CALCULATION AND EVALUATION OF INDICATORS FOR THE USE OF FIXED CAPITAL 9

1.3. CALCULATION AND EVALUATION OF THE EFFICIENCY OF THE ENTERPRISE USING INDEX INDICATORS 12

CHAPTER II. ANALYSIS OF PROFITABILITY INDICATORS 14

CHAPTER III. STATISTICAL ANALYSIS OF THE FINANCIAL ACTIVITY OF THE ENTERPRISE 22

CONCLUSION 30

LIST OF REFERENCES USED 31

LIST OF REFERENCES USED

1. Ayvazyan S.A., Mkhitaryan V.S. Applied Statistics and fundamentals of econometrics: Textbook for universities. – M.: UNITY, 1998.

2. Bakanov M.I., Sheremet A.D. Theory economic analysis: Textbook. – M.: Finance and Statistics, 1997.

3. Civil Code Russian Federation Part 1. – M.: Prospekt, 1997.

4. Gusarov V.M. Statistics theory: Tutorial for universities. – M.: Audit, UNITY, 1998.

5. Eliseeva I.I. General theory of statistics: Textbook for universities. – M.: Finance and Statistics, 1999.

6. Efimova M.R. Workshop on general theory Statistics: Textbook. – M.: Finance and Statistics, 1999.

7. Kotler F. Fundamentals of Marketing: Transl. from English – M.: Progress, 1990.

8. Maksimov O. B. Analysis of the financial state of the enterprise. Basic provisions of the methodology. – St. Petersburg: IKF “ALT”, 1994.

9. Mkhitaryan V.S. Statistics. – M.: Economist, 2005.

10. Decree of the Government of the Russian Federation of May 20, 1994 No. 498 “On some measures to implement legislation on the insolvency (bankruptcy) of enterprises.”

11. Russian statistical yearbook: Stat. Sat./Goskomstat of Russia. – M.:, 2001.

12. Ryabushkin B.T. Fundamentals of financial statistics: Textbook. – M.: Finstatinform, 1997.

13. Financial statistics: Textbook / Ed. Prof. V.N. Salina. – M.: Finance and Statistics, 2000.

15. Financial Economics/ Ed. Yu.M. Osipova, V.G. Belolipetsky, E.S. Zotova. – M.: Yurist, 2001.

1. Program of the discipline “Methods of applied statistics and econometrics” (adaptation course) for direction 080100. 68 “Economics” of the master’s program

   Discipline program

   In his research work Economists have to analyze a variety of data. Properly used statistical methods of data analysis significantly expand the possibilities of scientific research.

2. The program of the discipline "Econometrics-2" for the direction

   Discipline program

3. Sample program name of discipline Econometrics Recommended for training direction 080200 “Management”

   Sample program

   The purpose of the discipline “Econometrics” is to train students in the methodology and technique of constructing and applying econometric models to analyze the state and assess the prospects for the development of economic and social systems in an interconnected environment

4. Program of the discipline “Computer methods for analyzing sociological data (introduction to mathematical statistics and data analysis)” for direction 040100. 68 “Sociology” for master’s preparation Government of the Russian Federation

   Discipline program

Year of manufacture: 2010

Genre: Econometrics

Publisher: Master

Format: PDF

Quality: OCR

Number of pages: 512

Description: The content of the textbook corresponds to current educational standards And curriculum higher educational institutions economic profile in the discipline "Econometrics". The peculiarity of this publication is that in its description traditional methods solutions to econometric problems are organically integrated for the first time (where this allows for increased accuracy and depth of analysis) modern methods multivariate statistical analysis, previously not included in the econometrics tools (in particular, discriminant and cluster analyses, principal component analysis, etc.).
The methods and models of regression analysis, binary and multiple choice, time series analysis may form the content of one or two core semester courses in econometrics as part of an undergraduate curriculum.
For undergraduates, graduate students, teachers, as well as specialists in applied economics and econometrics.

You are holding in your hands a textbook on the methods of econometrics - a discipline that is one of the three basic disciplines (along with micro- and macroeconomics) of higher education. economic education. Unfortunately, this status of econometrics in Russia was recognized very late: only starting in 1992, econometrics was introduced into the economic education curricula of some leading Russian universities. This late recognition of econometrics immediately put Russian students to a disadvantage: by that time, only a few relatively old translated books on econometrics had been published in Russia, and the first domestic textbooks on this discipline appeared only in 1997-1998. (see [Magnus, Katyshev, Peresetsky (2005)], [Ayvazyan (2001)]). However, now the situation has improved significantly: the mentioned two books have been republished many times, domestic textbooks edited by I.I. have been published. Eliseeva (2006), V.I. Suslova (2005), translations from English wonderful books[Berndt (2005)], [Magnus, Neidecker (2007)], [Verbeek (2008)].
The ability to use the best examples of English-language econometric literature has significantly increased (due to increased general level possessions English our students and specialists, as well as the development of electronic communications, see, for example, the list of English-language literature at the end of this publication).
Under such circumstances, a natural question arises: what prompted the author to create another textbook on econometrics?
To answer this question, first of all, I must note that my understanding of the nature and purpose of econometric methods differs somewhat from that generally accepted in the North American and Western European econometric community. This understanding was formed on the basis of theoretical-probabilistic and mathematical-statistical national school in the process of getting acquainted with the best examples of English-language econometric literature, as well as personal scientific contacts with colleagues from Harvard University(USA), University of Paris 1/Sorbonne (France), Tilburg and Rotterdam Universities (Holland), University of Geneva (Switzerland) and other educational and scientific centers peace. The essence of these differences is briefly presented in paragraphs. 1.1 and 1.2 of Chapter 1 (Introduction) of the book. It should be added that over time, specialists’ ideas about the range of econometrics methods are somewhat transformed, and the emphasis in assessing the areas of their application shifts. I cannot agree with all such ideas accepted, say, in scientific circles in the United States. For example, it is customary to include in courses (textbooks) on econometrics “The Theory of Large Samples” (or “Asymptotic Theory”), “Nonparametric and Semiparametric Methods of Acceptance statistical solutions", a detailed presentation of the method maximum likelihood. But all these topics are traditionally presented as sections in other independent scientific disciplines— probability theory and mathematical statistics. At the same time, the most important applied methods of multivariate statistics for econometric analysis (discriminant and cluster analyses, principal component analysis, etc.) are, for unknown reasons, absent from econometric courses and classic university textbooks North America And Western Europe. I will add to this that over the past few years, some special methods multivariate statistical analysis, a number of important results in the field of financial econometrics, used in econometric analysis of financial data in risk management problems.
All the mentioned circumstances determined the specific differences between this publication and traditional textbooks on econometrics. Among these differences, first of all, one should highlight the fact that for the first time, as far as I know, in the description of traditional methods for solving econometric problems, procedures of multivariate statistical analysis that were not previously taken into account (such such as cluster analysis, discriminant analysis, principal component method).
Among the features of the book is the fact that it includes two extensive introductory chapters on regression (Chapter 2) and correlation (Chapter 3) analyzes. Many years of teaching practice in leading Russian universities(Moscow School of Economics, M.V. Lomonosov Moscow State University, Faculty of Economics MSU, State University - High school Economics, Russian Economic School, Moscow State University of Economics, Statistics and Informatics) convinced me that when starting to master econometrics, students, as a rule, have a clear lack of knowledge and skills in the basics of these two sections.
Returning to the question of the motivation for creating the textbook, it should be recognized that many years of research and pedagogical work author at the Moscow School of Economics, Moscow state university them. M.V. Lomonosov. Without constant working contacts with colleagues in the Department of Econometrics and mathematical methods economics, without the main critics and generators of questions - students of the Moscow School of Economics at Moscow State University, this book would hardly have been published.
The textbook covers a very complete range of methods of mathematical and statistical tools of econometrics in all its traditional sections, namely:

1. classical linear regression model and classic method least squares(chapters 4 and 6);
2. generalized linear regression model and generalized least squares method (Chapters 5 and 6);
3. linear regression model with variable structure (Chapter 8);
4. regression models with a discrete dependent variable: binary and multiple choice models (Chapter 9);
5. regression models under conditions of censoring, truncation or the sample selection of the dependent variable (Chapter 9);
6. statistical analysis univariate and multivariate time series (Chapter 10).

The appendices contain, in addition to mathematical and statistical tables, information from matrix algebra and multivariate statistical analysis. It is assumed that the reader already has the necessary training in mathematical statistics within the framework of basic courses provided state standards For economic specialties universities
It should, however, be emphasized that the proposed publication, of course, does not present all the most important sections of modern econometrics. It does not, for example, contain methods and models for analyzing multivariate time series, analyzing panel data, or the generalized method of moments; they do not reflect latest achievements in the field of financial econometrics (copula functions, financial risk management methods), the Bayesian approach to econometric analysis and methods for measuring and analyzing synthetic latent categories that comprehensively characterize the quality or efficiency of the functioning of the analyzed system. All these issues will be presented in an advanced econometrics course (intended for master's level of economic education), which is being prepared for publication by me and my Italian colleague (at the Moscow School of Economics) Dean Fantazzini.
As for the basic bachelor's level of the discipline "Econometrics", it is provided with the methods and models presented in this textbook, which can make up the content of one or two (depending on the area allocated in the course). curriculum university time) semester courses according to the scheme: 2 hours of lectures and 2 hours seminars per week. These classes should, of course, be equipped with tasks and exercises (including in computer classes), for which, in addition to the examples given in the book, we can recommend, for example, “Collection of problems for initial course econometrics” P.K. Katysheva, Ya.R. Magnus and A.A. Peresetsky (Delo publishing house, 2008).
The computational implementation of the methods described in the book is based on the use of statistical and econometric packages SPSS, E-views, R and STATA.
The author tried to follow a style of presentation that would help the reader understand, first of all, the main idea and meaning of the described method, and avoid a purely formal, mechanistic perception of the material. True, this is inevitably associated with an increase in the volume of the book.
In conclusion, I want to express my gratitude. First of all, I am grateful to the teams and administration of the MSE MSU and the Central Institute of Economics and Mathematics of the Russian Academy of Sciences, whose fruitful professional environment significantly helped in the work on the textbook. I received great benefit from communicating with colleagues - teachers of econometrics and statistics various universities Russia, Lithuania, Moldova during a specially organized series of seminars (1997-2002), at which domestic and foreign experts (including the author of the textbook) presented their lecture series within general program advanced training. Finally, I am grateful to Alla Pavlovna and Galina Yuryevna Grokhotov for their dedicated and professional work in preparing the original layout of the book.
I would like to draw the reader's attention to the following fact: to achieve success in the applications of econometric methods, the econometrician has to very delicately balance between economic theory, the possibilities of the necessary information support, the formulation of the initial assumptions of the model and the methods themselves. In other words, applied econometrics is not only a science, but also an art, mastery of which is learned through experience. So I wish the reader success in comprehending the science and art of mastering the subtle and effective tools of econometrics!

"Methods of econometrics"

Introduction

1. Econometrics: Evolution of Definition and Reality
2. Impoverishment of the mathematical apparatus of econometrics
3. The place of econometrics among mathematical, statistical and economic disciplines
4. Econometric model and problems of econometric modeling

Introduction to Regression Analysis

1. General formulation of the problem statistical research dependencies
2. What is the ultimate applied goal of statistical dependency research?
3. Some typical tasks econometric modeling practices
4. Main types of dependencies between quantitative variables
5. About the choice general view regression functions

Introduction to Correlation Analysis

1. Purpose and place of correlation analysis in statistical research
2. Correlation analysis quantitative characteristics
3. Correlation analysis of rank (ordinal) variables: rank correlation
4. Correlation analysis of categorized variables: contingency tables

Classical linear multiple regression model (CLMRM)

1. Description of KLMMR. Basic assumptions of the model
2. Assessment unknown parameters KLMMR: least squares and maximum likelihood method
3. Analysis of the variation of the resulting indicator in and sampling factor determination
4. Multicollinearity and selection of the most significant explanatory variables in KLMMR
5. KLMMR with linear constraints on parameters
6. A general approach to statistical testing of hypotheses about the presence linear connections between KLMMR parameters

Generalized linear multiple regression model

1. Description of the generalized linear model multiple regression(OLMMR)
2. Estimates of GLMMR parameters using the generalized least squares method (GLS-estimates)
3. GLMMR with heteroscedastic residuals
4. GLMMR with autocorrelated residuals
5. Practical implementation of OMC ( general approach)

Forecasting based on linear multiple regression models

1. Analysis of the accuracy of the estimated LMMR (theoretical basis for solving forecast problems)
2. Best point forecast y(X) and f(X) = E(y|X) based on OLMMR
3. Interval forecast y(X) and f(X) = E(y|X), based on OLMMR
4. Analysis of regression model accuracy and forecasting in a realistic situation

Linear regression models with stochastic explanatory variables

1. Random residuals e do not depend on predictors X and estimated regression coefficients
2. General case: Stochastic predictors X are correlated with regression residuals. Instrumental Variables Method
3. Random errors in measuring the values ​​of explanatory variables

Linear regression models with variable structure

1. The problem of heterogeneous (in the regression sense) data
2. Introducing dummies (dummy variables) into a linear regression model
3. Testing regression homogeneity of two groups of observations (G. Chow test)
4. Construction of KLMMR from heterogeneous data in conditions when the values ​​of associated variables are unknown

Models with discrete and discrete-continuous dependent variables

1. Binary choice models
2. Multiple Choice Models
3. Relationship between binary and multiple choice models and discriminant analysis
4. Model with discrete-continuous dependent variable (Tobit model)

Univariate time series analysis (models and forecasting)

1. Time series: definitions, examples, formulation of main tasks
2. Stationary time series and their main characteristics
3. Non-random component of a time series and methods for smoothing it
4. Models of stationary time series and their identification
5. Models of non-stationary time series and their identification
6. Forecasting economic indicators, based on the use of time series models

Appendix 1. Tables of mathematical statistics
Appendix 2. Required information from matrix algebra
Appendix 3. Multivariate statistical analysis
Literature

Econometric methods. Ayvazyan S.A.

M.: 2010 - 512 p.

The content of the textbook corresponds to the current educational standards and curricula of higher educational institutions of economics in the discipline “Econometrics”. The peculiarity of this publication is that for the first time, modern methods of multivariate statistical analysis, which were not previously included in the econometrics tools (in particular, discriminant and cluster analyses, principal component analysis, etc.). The methods and models of regression analysis, binary and multiple choice, and time series analysis presented in the textbook can form the content of one or two basic semester courses in econometrics as part of the undergraduate curriculum. For undergraduates, graduate students, teachers, as well as specialists in applied economics and econometrics.

Format: pdf

Size: 5.8 MB

Download: Rghost

TABLE OF CONTENTS
Preface 9
Chapter 1. Introduction 13
1.1. Econometrics: Evolution of Definition and Reality 13
1.2. Impoverishment of the mathematical apparatus of econometrics 16
1.3. The place of econometrics among mathematical, statistical and economic disciplines 19
1.4. Econometric model and problems of econometric modeling 22
Conclusions 30
Chapter 2. Introduction to regression analysis 33
2.1. General formulation of the problem of statistical research of dependencies 33
2.2. What is the ultimate applied goal of statistical dependency research? 42
2.3. Some typical tasks of econometric modeling practice 45
2.4. Basic types of dependencies between quantitative variables 50
2.5. On choosing the general form of the regression function 55
Conclusions 65
Chapter 3. Introduction to Correlation Analysis 67
3.1. Purpose and place of correlation analysis in statistical research 67
3.2. Correlation analysis of quantitative characteristics 69
3.3. Correlation analysis of rank (ordinal) variables: rank correlation 96
3.4. Correlation analysis of categorized variables: contingency tables 111
Conclusions 117
Chapter 4. Classical linear multiple regression model (CLMMR) 121
4.1. Description of KLMMR. Basic assumptions of the model 121
4.2. Estimation of unknown KLMMR parameters: least squares method and maximum likelihood method 126
4.3. Analysis of the variation of the resulting indicator y and the sample determination coefficient Shx 140
4.4. Multicollinearity and selection of the most significant explanatory variables in KLMMR 145
4.5. KLMMR with linear constraints on parameters 162
4.6. General approach to statistical testing of hypotheses about the presence of linear relationships between CLMMR parameters 167
Conclusions 176
Chapter 5. Generalized Linear Multiple Regression Model 179
5.1. Description of the generalized linear multiple regression model (GLMMR) 179
5.2. Estimates of GLMMR parameters using the generalized least squares method (GLS-estimates) 183
5.3. GLMMR with heteroscedastic residuals 188
5.4. GLMMR with autocorrelated residuals 198
5.5. Practical realizable OMNC (general approach) 207
Conclusions 210
Chapter 6. Forecasting Based on Linear Multiple Regression Models 213
6.1. Analysis of the accuracy of the estimated LMMR (theoretical basis for solving forecast problems) 214
6.2. Best point forecast y(X) and f(X) = E(y|X) based on OLMMR 216
6.3. Interval forecast y(X) and f(X) = E(y|X), based on OLMMR 220
6.4. Analysis of the accuracy of the regression model and forecasting in a realistic situation 226
Conclusions 230
Chapter 7. Linear regression models with stochastic explanatory variables 233
7.1. The random residuals e are independent of the predictors X and the estimated regression coefficients in 235
7.2. General case: stochastic predictors X are correlated with regression residuals e. Instrumental variable method 238
7.3. Random errors in measuring the values ​​of explanatory variables 243
Conclusions 249
Chapter 8. Linear regression models with variable structure 251
8.1. The problem of heterogeneous (in the regression sense) data 251
8.2. Introducing “dummies” (dummy variables) into a linear regression model 254
8.3. Checking the regression homogeneity of two groups of observations (G. Chow test) 263
8.4. Construction of KLMMR using heterogeneous data in conditions where the values ​​of associated variables are unknown 265
Conclusions 269
Chapter 9. Models with Discrete and Discrete-Continuous Dependent Variables 271
9.1. Binary choice models 273
9.2. Multiple Choice Models 282
9.3. Relationship between binary and multiple choice models and discriminant analysis 285
9.4. Model with a discrete-continuous dependent variable (Tobit model) 287
Conclusions 291
Chapter 10. Univariate Time Series Analysis (Models and Forecasting) 293
10.1. Time series: definitions, examples, formulation of main tasks 295
10.2. Stationary time series and their main characteristics 302
10.3. Non-random component of a time series and methods for smoothing it 314
10.4. Models of stationary time series and their identification 336
10.5. Models of non-stationary time series and their identification 378
10.6. Forecasting economic indicators based on the use of time series models 395
Conclusions 409
Appendix 1. Tables of mathematical statistics 413
Appendix 2. Necessary information from matrix algebra.. 433
Appendix 3. Multivariate statistical analysis 455
Literature 493
Alphabetical subject index 497