STATISTICAL MODELLING_BA_2015

advertisement

Правительство Российской Федерации

Федеральное государственное автономное образовательное учреждение высшего профессионального образования

"Национальный исследовательский университет

"Высшая школа экономики"

Факультет экономических наук

Программа дисциплины

« Статистическое моделирование социально-экономических процессов»

(Statistical Modelling of Social and Economic Processes)

Для направления 38.03.01. «Экономика» подготовки бакалавра

Авторы программы:

Сиротин В.П., vsirotin@hse.ru

Одобрена на заседании

Рекомендована секцией УМС

Председатель В.С. Автономов

« » 20__г.

« » 20__г.

Ученый секретарь Т.В. Коссова

Москва, 2016

Настоящая программа не может быть использована другими подразделениями университета и другими вузами без разрешения кафедры-разработчика программы.

Национальный исследовательский университет «Высшая школа экономики»

Программа дисциплины «Статистическое моделирование социально-экономических процессов» (Statistical modelling of social and economic processes) для направления 38.03.01. «Экономика» подготовки бакалавра

Statistical Modelling of Social and

Economic Processes

Syllabus

Jan, 2016 Faculty: Economics

Year: 2015/16

Course name: Statistical Modelling of Social and Economic Processes

Level: Bachelor, 4 th year

Language of instruction: English

Period: Module 3

Course instructor and contact details

Viacheslav Sirotin (lectures & classes), vsirotin@hse.ru

Course description

Methodology of statistical analysis of social and economic processes is in the focus of the course.

Among other processes, the attention is paid to differentiation of the population on income, typology of consumers, economic and innovative development of regions and countries, quality of life. The course is aimed to creating relevant model of the process, starting with theoretical foundation. The first step of modelling also includes gathering various types of qualitative and quantitative data that may be presented in either fixed or fuzzy form according to the nature of variables involved. The key point here is transformation of initially interconnected data into the index form along with dimensionality reduction using either eigenvalues or singular decomposition of the data matrix.

Decomposition of the probability density function is considered to be the main way of the structure analysis in a low dimensional space; the results of the decomposition are used for soft and crisp classifications. When clustering, we can consider pros and cons of the ordinary “k-means” method and its fuzzy modifications. These and other methods are used for creating the models. All the models should be identified using modern computer software.

Course objectives

The course is oriented on the problems of statistical practice in creating models of social and economic processes. Students will be familiar with the features of expert information and the ways of its using along with the data of more precise measuring. Students will be able to create and analyze structural models of social and economic systems on the base of crisp and fuzzy approach. Students will be able to create integral indicators in the cases of presence and absence of learning.

2

Национальный исследовательский университет «Высшая школа экономики»

Программа дисциплины «Статистическое моделирование социально-экономических процессов» (Statistical modelling of social and economic processes) для направления 38.03.01. «Экономика» подготовки бакалавра

Forms of study and control

The course consists of lectures and classes. Each lecture is followed by a class where students solve numeric problems. The course requires a sufficient amount of self-study. The final grade is determined by class participation, home assignments, and a written final exam in the end of the course. The final exam is 60 minutes long.

Grade determination

25% - class participation, 50% - home assignments, 25% - final exam.

Textbooks

Scott DW (1992) Multivariate density estimation: theory, practice, and visualization. Wiley, New

York

Dubois D., Prade H. (2000) Fundamentals of Fuzzy Sets. – Boston: Kluwer Academic Publishers

Course outline

1.

Statistical approach to social and economic modelling

The main stages in creating statistical models of socio-economic processes. Information support for modeling. Scaling of the data. Using expert information for statistical modelling. Constructing of the latent variables. Fuzzy-set approach to enhance the scale level of measurement. The problem of missing data.

2.

Statistical models of social stratification

Probability density estimation in exploratory data analysis. Parametric approach to mixture density estimation Classification based on decomposition of probability density function. Crisp and fuzzy strata borders. Estimation of the probability density function on the base of interview data..

3.

Crisp and soft clustering of social and economic objects

Types of social and economic structures. "k-means" and "c-means" clustering. Membership function in crisp and fuzzy clustering. Evolution of the socio-economic structure. Using membership function in class description.

3

Национальный исследовательский университет «Высшая школа экономики»

Программа дисциплины «Статистическое моделирование социально-экономических процессов» (Statistical modelling of social and economic processes) для направления 38.03.01. «Экономика» подготовки бакалавра

4.

Index constructing

Expert-statistical technique in index constructing. Hierarchical system of integral indicators.

Examples of the integral indicators systems.

Distribution of workload

# Topic TOTAL

(hours)

36

Audience hours Homework including:

(hours)

Lectures Classes

4 4 28 1 Statistical approach to social and economic modelling

2 Statistical models of social stratification

3 Crisp and soft clustering of social and economic objects

4 Index constructing

TOTAL:

36

36

36

144

4

4

4

16

4

4

4

16

28

28

28

112

4

Download