EC 531 Term Paper Suat AKBULUT
EC 531 ECONOMETRICS I
Term Paper
Suat AKBULUT
2013749003
In my proposal i wrote that i would study to explain the deviations in
Gini coefficient, however, due to some unlucky data finding difficulties and the uninteresting results i have decided to change my term paper subject, with the course instructor’s consent, to a model which works with less sophisticated data. Yet, it might be even benevolent to start applied econometrics with such a model.
The Aim of the Paper and Regression Model
This paper aims to understand the factors that affect the number of crimes in 1993 in each state of the U.S, and to analyze how they affect the number of crimes. In order to do this, it uses a lin-lin regression model because of the advise that the plotted data gives. Following is the regression model the
1
paper uses to explain the deviations in the number of crimes,
crime = β0 + β1 · pov + β2 · metro + β3 · popdens +
where crime refers to total number of crimes comitted in 1993 by each state in the U.S, pov stands for percentage of the population living below the poverty level in 1993 by each state, metro is used for metropolitan population as percentage of the state population and popdens is the population per square mile by each state in the same year.
Expectations
Firstly, when one looks at the first independent variable (pov), it is realistic that she could assume that the increase in the percentage of people who lives below the poverty level would lead to an increase in the dependent variable (crime), the number of total crime. It is because people not able to earn sufficient money to maintain their life might become more prone to illegal activities such as burglary, murdering etc. Furthermore, people living below the poverty level might also be under-educated, which also increases the probability of these people to do illegal activities. Hence omitting such a variable might cause the model to have omitted variable bias, yet, lack of data does force the model to run
Term Paper
Suat AKBULUT
2013749003
In my proposal i wrote that i would study to explain the deviations in
Gini coefficient, however, due to some unlucky data finding difficulties and the uninteresting results i have decided to change my term paper subject, with the course instructor’s consent, to a model which works with less sophisticated data. Yet, it might be even benevolent to start applied econometrics with such a model.
The Aim of the Paper and Regression Model
This paper aims to understand the factors that affect the number of crimes in 1993 in each state of the U.S, and to analyze how they affect the number of crimes. In order to do this, it uses a lin-lin regression model because of the advise that the plotted data gives. Following is the regression model the
1
paper uses to explain the deviations in the number of crimes,
crime = β0 + β1 · pov + β2 · metro + β3 · popdens +
where crime refers to total number of crimes comitted in 1993 by each state in the U.S, pov stands for percentage of the population living below the poverty level in 1993 by each state, metro is used for metropolitan population as percentage of the state population and popdens is the population per square mile by each state in the same year.
Expectations
Firstly, when one looks at the first independent variable (pov), it is realistic that she could assume that the increase in the percentage of people who lives below the poverty level would lead to an increase in the dependent variable (crime), the number of total crime. It is because people not able to earn sufficient money to maintain their life might become more prone to illegal activities such as burglary, murdering etc. Furthermore, people living below the poverty level might also be under-educated, which also increases the probability of these people to do illegal activities. Hence omitting such a variable might cause the model to have omitted variable bias, yet, lack of data does force the model to run