Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
211 views

Structure Project Topics For Projects 2016

This document outlines the requirements for econometric projects to be completed by Prof. Dr. D. Jemna in 2016-2017. It provides guidance on acceptable domains and subjects, data sources, required project structure, and statistical analysis. Projects must use quantitative cross-sectional data from at least 30 statistical units to build a multiple regression model for an output variable. The project structure involves introducing the topic, presenting relevant theory and hypotheses, collecting data, descriptive statistics, single regression models, a final multiple regression model, and testing assumptions.

Uploaded by

Mariu Vornicu
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
211 views

Structure Project Topics For Projects 2016

This document outlines the requirements for econometric projects to be completed by Prof. Dr. D. Jemna in 2016-2017. It provides guidance on acceptable domains and subjects, data sources, required project structure, and statistical analysis. Projects must use quantitative cross-sectional data from at least 30 statistical units to build a multiple regression model for an output variable. The project structure involves introducing the topic, presenting relevant theory and hypotheses, collecting data, descriptive statistics, single regression models, a final multiple regression model, and testing assumptions.

Uploaded by

Mariu Vornicu
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Prof.dr. D.

Jemna, Econometric Projects, 2016-2017


Domain
1. Macro-economics

2. Business

3. Demography

4. Social

5. Health
6. Culture

7. Sport
8. Environment

9. Tourism
10. Transport

Subjects
-

Consumption
Inflation, Interests
Income
Labor market
Industry, Agriculture, trades
Government
Entrepreneurship, New companies
Transparency
Profits, wages
Infant mortality
Fertility, Marriage
Migration
Unemployment, Poverty
Corruption, Criminality
Education
Health status, Health care
Diseases, AIDS
Movies, cinema industry
Participation, Expenditure
Artists
Participation
Expenditure
Pollution
Energy
Waste
Capacity
Occupancy
Roads
Railway

Observations
- Population or sample should be more than 30 units (countries, regions, counties, companies, peoples etc.)
- Possible populations: EU countries and candidates 28+some other countries; Romanian counties (42 or 41 less Bucharest); a sample of
European regions (Eastern Europe regions, central Europe regions, occidental Europe regions etc.); a sample of worldwide countries (selected
by regions or economic developments etc.)
- GDP/cap should be an independent variable
- All variables must be quantitative
- As independent variables minimum 5 factors
- All variables must be cross-sectional data (for a certain moment in time: year, month, day), but not times series
- Data sources:
http://databank.worldbank.org/data/home.aspx
www.insse.ro
http://ec.europa.eu/eurostat/data/database
http://www.un.org/en/development/desa/population/
http://data.unicef.org/
http://www.transmonee.org/
http://www.doingbusiness.org/data

Project structure
1. Introduction
- Outline for the project subject (present the problem, the concept or phenomenon, importance, actuality, context)
- Define the population and statistical units
- Define the objective of the project: to build an econometric model in order to identify the explanatory factors for the main phenomenon (we
call it output)
An overview on literature can help here (see the scientific articles from internet)
Example 1 Book, p.4, consumption
Example 2 infant mortality it is an important demographic phenomenon that represent an important issue for EU countries etc.
Literature - Jemna, 2014, Infant mortality and the socioeconomic conditions in the CEE countries after 1990
Population EU 28 countries plus 5 candidates (Turkey, Macedonia, Montenegro, Serbia, Albania).

2. Statement of theory
- Using the scientific literature, present a short theory for the selected subject of the project (a set of ideas that explain the output or the
problem formulated in the introduction)
- Definitions, explanations, approaches
- Identify the possible explanatory factors (economic, social, demographic variables) for the output and the possible type of relationships
between the factors and output
- Define some hypotheses which can be tested with the help of econometric modeling procedure (assumptions about the relationship between
each factor and the output)
Example 1 Book, p.4, Keynes theory; 2 variables: consumption and income
Example 2 for infant mortality there is a huge literature in which we can identify some theoretical assumptions, some explanatory factors
(economic, social and demographic factors) Again, see the paper Jemna, 2014.
A hypothesis: there is a strong and negative relationship between infant mortality and economic development for EU countries
Variables: infant mortality, economic development, unemployment, income, fertility, marriage, age structure, health system, level of
urbanization etc.

3. Collecting data
- Define the statistical variables and the measurement units (dependent and independent variables)
- Define the database (in excel or spss)
- Download the statistical data from the official source
Example 1 Book, p.6, dependent variable is personal consumption expenditure (bill. dollars) and the independent variable is GDP (bill.
dollars)
Example 2 dependent variable is infant mortality rate (deaths under 1 year to 1000 persons); independent variables: GDP/capita ($),
unemployment rate (%), fertility rate (no of children per women), average income ($) etc.
Data are downloaded from http://ec.europa.eu/eurostat/data/database

4. Descriptive statistics for data


- Using SPSS or R software, for each variable are presented the basic numeric indicators and the visual presentation of each distribution
- The goal of this step is to evaluate the nature of the statistical data (homogeneity, possible outliers etc.)
- In this step we can decide to transform some variables or to eliminate some units from the database
Example in SPSS, the menu Analyze/Descriptive Statistics/Explore/ (Statistics Descriptives and Outliers; Plots Boxplot, Histogram and
Normality plot)
5. Single equation regression
- Create an econometric linear model between each independent variable and the dependent variable
- The purpose of this step is to evaluate the type and significance of each relationship in order to prepare the way for the multiple regression
model
- We can decide for a certain variable to be transformed with log function if we conclude that exist a nonlinear relationship with the output
- For each model are presented the scatter-plot, and the 3 tables with statistical results: Coefficients, Anova, Model summary
Example in SPSS, the menu Analyze/Graphs/Legacy Dialogs/ Scatter/Dot
The menu Analyze/Regression/Linear/
6. Multiple regression model
- Create a multiple regression model (linear or not) with all independent variable
- Exclude the irrelevant independent variables
- Rank the importance of significant factors, estimate the correlation indicators
- Test the hypotheses formulated in the second step
Example in SPSS, the menu Analyze/Regression/Linear/ (Statistics R squared Change, Partial and Part correlations, Method Backward)
7. Testing classical assumptions
- The model estimated in the last step must be tested for several classical assumptions. For any problem, the model can be corrected.
8. Conclusions
- Some remarks about the estimated econometric model (significant factors, possibilities to use the model for decisions etc.)
9. References

You might also like