just only want steps from 6 to 10 and example attached

GDP GROWTH ON MORTALITY RATE IN

2

000 AND 201

4

ACROSS THE WORLD

Yuxuan Tang

Jianing Wang

EC204 Empirical Economics II, Fall 202

1

ABSTRACT

Journal of Population of Economics stated “For the period 1

8

00–2000, an increase in

GDP by 1% decreased mortality by 0.

7

%. This overall relationship is due to a strong

counter-cyclical relationship in the nineteenth century, which disappeared in the twentieth

century” (Svensson, M., Krüger, 20

10

). Based on the WorldBankData2years panel data in the

year of 2000 and 20

14

, this research mainly focused on the effects of GDP growth on mortality

rate, with different variables involved. The results showed a statistically insignificant relationship

between GDP growth and mortality rate. And 4

3

.3

9

% of the variation of the mortality rate can be

explained by GDP growth within the country when holding country level and time fixed effect.

1

⻄

https://learn.bu.edu/bbcswebdav/pid-9

5

75

6

30-dt-content-rid-590

11

850_1/xid-59011850_1

I. INTRODUCTION

This research aims to discover if GDP per capita affects mortality rate. Past research

shows that GDP per capita is inversely related to mortality rate during 1901-2000 in the United

States (M Harvey Brenner, 2005). In this article, Thomas McKeown demonstrated that

economic development is of fundamental importance to the decline of classic infectious and

childhood disease. With rapid economic growth in the 20th century, more people tend to have

vaccinations and are less vulnerable to infectious and childhood disease, which leads to a

decline in mortality rate. As a result, an inverse relationship between GDP per capita and

mortality rate worldwide was expected at the beginning.

After the hypothesis was conducted, we described and utilized a panel data across the

world in 2000 and 2014, and regressed GDP per capita and mortality rate with some variables

including improved sanitation facilities of po, co2 emissions metric tons per capita, improved

water, urban population growth annual spurb, prevalence of hiv total of population,

immunization measles of children age and others are tested with GDP per capita to find out how

it affects mortality rate. Then, we compiled our findings and found there is a statistically

insignificant relationship between the two main variables. Therefore, we use interactive variables

to test if the effects of GDP growth per capita on mortality rate depends on other variables listed

above. Then we created a graph that involves a linear regression and scatter plot were used to

make further comparison of fitness. Also, with the quadratic model being graphed, the turning

point is at 0.

16

4310932, and after this turning point, the relationship between GDP growth and

mortality rate becomes positive contrary to our expectations.

2

2. Literature Review

Many researchers had done studies relative to the effects of GDP growth on mortality rate

for years, and the reasons could be complicated. Mikael Svensson and Niclas A. Kruger used

wavelet methods to analyze the relationship between mortality rate and economic growth from

1800 to 2000 in Sweden. (Mikael Svensson and Niclas A. Krüger, 20

12

) According to the article,

it was found that in the early period of the 19th century, people were more vulnerable to disease

and health problems when the economy went downward. As a result, the mortality rate was

higher when the economy was poor. However, when we entered the 20th century, the augment

changed. People were more likely to stress out due to reasons including work stress, family

pressure due to unemployment, which leads to higher death rate. Furthermore, the research found

out some more specific factors that associate mortality rate with GDP growth, including stroke,

accident, suicide, cancer, and infection.

More findings were found by M Harvey Brenner. Using the time series model, with

variables of “ long-term effects of economic growth over 0–11 years,” “long-term effects of

unemployment over 0–11 years,” and “interactive effect of unemployment and GDP per capita

over 0–11 years”, it was found out that for a short period, increased mortality rate was due to

higher GDP growth, because of better technology with longer working period and speed.

However, for a longer period, GDP growth leads to the decline of mortality rate.（M Harvey

Brenner, 2005） More evidence was found by Brenner and Haines to prove this theory. According

to the article written by Haines in 2003, it was found that the United States experienced a rapid

economic growth but rising mortality rate between 1830 and 1860 due to deterioration of the

biological standard of living (Hanis 2003). During this period, the fast urban growth, mass

migration from abroad, changes in transportation infrastructure, rapid commercialization,

3

worsened the mortality environment which caused the mortality rate to rise. For a longer period,

Varvarigos constructed a model of a growing economy with pollution and testified that economic

growth and mortality rates are negatively related due to the difference of environment-related

structural parameters, such as lower p (units of pollution per output generating), which improves

the environmental conditions and reduces mortality rate (Varvarigos, 20

13

).

3. Data Description

Table 1

This research used panel data at country level worldwidely in the year of 2000 and 2014

from world bank data to analyze the relationship between GDP growth and mortality rate. A total

of 369 observations are collected from world bank data with 6 variables, including sanitation

facilities of po, co2 emissions metric tons per capita, improved water, urban population growth

annual spurb, prevalence of hiv total of population, and immunization measles of children age.

These 6 variables, together the two main variables are tested to find out the relationship between

GDP growth and mortality rate. The six variables are chosen because we realized that the higher

GDP a country has, the more conscious people have of their health. And as a result, more people

4

are getting vaccinated and actions or policies are taken for the sake of citizens’ health, which

leads to the decline of mortality rate.

The data of this research all come from world bank data, and two tables were created by

different years to describe the mean and standard deviation of the variables. Out of all the

variables, improved water has the highest mean value of 83.2% and 89.0 % in 2000 and 2014,

whereas urban population growth annual spurb have the lowest mean values around 2% in both

years.

Table 2

5

4. Model:

After we collected the data, we constructed a model of mortality rate as a function of GDP

growth at the country level of time fixed effect.

Within this fixed effect model, by holding year t and country i at constant level, mortrate

represents mortality rate, the continuous dependent variable in this equation, in year t and

country i. The main independent variable of this equation is gdpgrowth, which is continuous in

country i and year t, and is predicted to have a positive relationship with the main variable

mortality rate. The model is predicted as a linear regression as shown in the scatterplot graph. As

we used a time fixed effect model, the 6 other variables with i are absorbed into the ai variable

which change based on different countries. According to graphs shown below, most countries

with different mortality rates are scattered between 0% to 20% growth of GDP in both 2000 and

2014.

Also, in this model, the panel data at country level analyzes data from both year of 2000

and year of 2014 by using the dummy variable d00t and u is the error term. Graph1 represents

the worldwide GDP growth rate and mortality rate in 2000, and graph 2 displays GDP growth

rate and mortality rate in both the years of 2000 and 2014. However, by looking at the two

graphs below, we can see there is no inverse relationship between the GDP growth rate and

mortality rate, but instead a positive relationship. However, we cannot conclude that there is a

definitely positive relationship between GDP growth rate and mortality rate, as the dots mostly

concentrated in the middle of the graph rather than displaying a linear relationship. And there are

6

countries including Liberia, Equatorial Guinea , and Timor-Leste which are more than 3 standard

deviations away from the mean fall into the category of becoming outliers of the group.

Therefore, we used some interactive variables to test if there is a non linear relationship between

the two main variables. (shown in table-3)

Graph1: Scatterplot of Worldwide GDP Growth Rate and Mortality Rate in 2000.

7

Graph 2: Two-way Scatterplot of Worldwide GDP Growth and Mortality rate in 2000

and 2014.

8

5.RESULTS

Table 3: Regression Results

Looking at table 3, the coefficient of GDP growth rate has a statistically insignificant

relationship with mortality rate, and we can not conclude that GDP growth rate has a linear

relationship with mortality rate. Therefore, we added 6 more variables as shown in Table 3 that

are relative to mortality rate to test their relationships. The results in Model 2 show that the

coefficient of improved sanitation facility sanitation and CO2 emissions are statistically

insignificant with mortality rate. And the coefficient of Immunization measles of children’s age,

prevalence of HIV total of population, and improved water are statistically significant at 1%

level with mortality rate, with P-value equals to 0. Urban population growth annual spurb is

statistically significant at 5% level on the country level, with P-value equals to 0.047. Therefore,

we removed these two insignificant variables and ran the regression (Model 3). Since it’s a panel

data, in order to make sure different countries have the same coefficient effect, we uses country

9

level fixed effects, as we can see in Model 4, the coefficient of GDP growth rate still has a

statistically insignificant relationship with mortality rate, even after we controlling for the effects

of time (Model 5).

Furthermore with the data, we decided to add interactive variables of immunization and

GDP growth rate in Model 6, within in a country and after controlling for the effects of year,

with variable we testified significance before , the data shows that the coefficient of GDP growth

rate still has a statistically significant relationship with mortality rate at 1% level because the

effect of GDP growth on mortality rate depends on the percentage of Kids Immunization (12-13

months), and 77.5% of the variable of data in mortality rate explained by GDP growth rate

within country when the effects of time controlled.

In Model 7, we tested if the relationship of GDP growth rate to mortality rate depends on

other 3 significance variables. The results showed that the other 3 coefficients of interactive

variables are statistically insignificant which does not affect the relationship of GDP growth rate

to mortality rate. The results shown are not as consistent with our hypothesis, as immunization is

the factor that would affect the relationship between the two main variables.

6. Conclusions

Based on our findings on the model, GDP growth does not have a statistically significant

relationship with the mortality rate. However, when holding countries and time fixed, and we

added the interactive variable immunization, the results showed a statistically significant

relationship between GDP growth rate and mortality rate. As a result, we can conclude that GDP

growth rate and mortality rate depends on a third variable of immunization measles of children’s

age. And our findings do not completely support the hypothesis that there is a linear relationship

10

between GDP growth rate and mortality rate. As shown in graph 1 and graph 2, the countries are

scattered altogether in a group instead of displaying a linear relationship. While the independent

variable being statistically insignificant, the causes can be complicated due to many other

factors.

7. Limitations and future results

For further research in the future, we would apply more different variables which could

contribute to the change of the dependent variable, including import and export on the country

level. Also, besides testing the interactive variables, maybe we would use the log and square of

the independent variable after we collected the data. And more research will be done not only on

a country level, but within a country as well. There are some limitations in this research,

including not enough variables that are relative to the dependent variable. More variables will be

applied in the future research.

11

REFERENCES CITED

● Svensson, M., Krüger, N.A. Mortality and economic fluctuations. J Popul Econ 25,

12

15

–1235 (2012). https://doi.org/10.1007/s00148-010-0342-8

● M Harvey Brenner, “Commentary: Economic growth is the basis of mortality rate decline

in the 20th century—experience of the United States 1901–2000”, International Journal

of Epidemiology, Volume 34, Issue 6, December 2005, Pages 1214–1221. Retrieved

from https://doi.org/10.1093/ije/dyi146

● Svensson, Mikael, and Niclas A. Krüger. “Mortality and Economic Fluctuations:

Evidence from Wavelet Analysis for Sweden 1800—2000.” Journal of Population

Economics, vol. 25, no. 4, Springer, 2012, pp. 1215–35,

http://www.jstor.org/stable/23354789.

● M Harvey Brenner, Commentary: Economic growth is the basis of mortality rate decline

in the 20th century—experience of the United States 1901–2000, International Journal of

Epidemiology, Volume 34, Issue 6, December 2005, Pages 1214–1221,

https://doi-org.ezproxy.bu.edu/10.1093/ije/dyi146

● Varvarigos. (2013). ENVIRONMENTAL DYNAMICS AND THE LINKS BETWEEN

GROWTH, VOLATILITY AND MORTALITY. Bulletin of Economic Research, 65(4),

314–331. https://doi.org/10.1111/j.1467-8586.2011.00410.x

● Haines, Craig, L. A., & Weiss, T. (2003). The Short and the Dead: Nutrition, Mortality,

and the “Antebellum Puzzle” in the United States. The Journal of Economic History,

63(2), 382–413. https://doi.org/10.1017/S0022050703001839

12

https://doi.org/10.1007/s00148-010-0342-8

https://doi.org/10.1093/ije/dyi146

http://www.jstor.org/stable/23354789

https://doi-org.ezproxy.bu.edu/10.1093/ije/dyi146

https://doi.org/10.1111/j.1467-8586.2011.00410.x

https://doi.org/10.1017/S0022050703001839

APPENDIX A

DO FILE

clear all

set more off

capture

log close

cd/Users/sarah

use “/Users/sarah/Downloads/WorldBankData2years (7).dta”

sum mortrate gdpgrowth immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

improvedsanitationfacilitiesofpo

co2emissionsmetrictonspercapitae

urbanpopulationgrowthannualspurb improvedwater

*Graph 2

bysort year: eststo: estpost sum mortrate gdpgrowth

immunizationmeaslesofchildrenage

prevalenceofhivtotalofpopulation improvedsanitationfacilitiesofpo

co2emissionsmetrictonspercapitae urbanpopulationgrowthannualspurb improvedwater

esttab using summary_stats_table.rtf, cells((mean(fmt(%10.2f)) sd(fmt(%10.2f)))) label

title(Summary Statistics) nonumber nomtitle replace

label var mortrate “Mortality Rate”

label var gdpgrowth “GDP Growth”

label var immunizationmeaslesofchildrenage “% of Kids Immunization (12-13 months)”

label var prevalenceofhivtotalofpopulation “HIV population”

label var improvedsanitationfacilitiesofpo “improved sanitation facility”

label var co2emissionsmetrictonspercapitae “CO2 emissions”

label var urbanpopulationgrowthannualspurb “Urban Population Growth”

13

label var improvedwater “Imporve Water”

#delimit

;

esttab using SummaryStats1 , main(mean) aux(sd)

nonotes rtf replace label varwidth(30) modelwidth(9) b(%9.2f) nonumbers

mtitle(“2000” “2014”)

title(“Table 1: Summary Statistics by Year”)

addnotes(“NOTE: Table reports the mean and standard deviation in 2000 and 2014. ‘The

mean’ is above ‘standard deviation’ for each variable”)

;

#delimit cr

*Graphing in 2000 and 2014

twoway (lfitci mortrate gdpgrowth )(scatter mortrate gdpgrowth), ytitle(Mortality Rate)

*title(Two-way Scatterplot of GDP Growth Rate and Mortality Rate)

*graph save “Graph” “/Users/sarah/Desktop/Two-way Scatterplot 2000 and 2014.gph”

*Graphing in 2000

twoway (scatter mortrate gdpgrowth if year == 2000, mlabel(countryname)) (lfit mortrate

gdpgrowth)

*ytitle(Mortality Rate) xtitle(GDP Growth)

*title(Two-way Scatterplot of GDP Growth and Mortality Rate)

*graph save “Graph” “/Users/sarah/Desktop/Scatterplot 2000.gph”

*Graph 3

*regress X and Y

reg mortrate gdpgrowth, r

14

outreg2 using ResearchRegression , replace label title (“Regression Results”) adjr2 addtext

(Country FE, NO, YEAR FE, NO)

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, NO, YEAR FE, No)

*regression on full model

reg mortrate gdpgrowth immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

improvedsanitationfacilitiesofpo co2emissionsmetrictonspercapitae

urbanpopulationgrowthannualspurb improvedwater,r

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, NO, YEAR FE, No)

*running an F test on the insignificant variables improvedsanitationfacilitiesofpo

co2emissionsmetrictonspercapitae

test gdpgrowth improvedsanitationfacilitiesofpo co2emissionsmetrictonspercapitae

*The F test shows these variables are jointly insignificance, so we kick these variable out

*To see if gdpgrowth is jointly significance with other variable, let’s pick

immunizationmeaslesofchildrenage

test gdpgrowth immunizationmeaslesofchildrenage

*regression on full model

reg mortrate gdpgrowth immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

urbanpopulationgrowthannualspurb improvedwater,r

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, NO, YEAR FE, No)

*Regression on Fixed Effect

xtset countrynum year

xtreg mortrate gdpgrowth, r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, No)

15

xtreg mortrate i.year gdpgrowth, r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year c.gdpgrowth##c.immunizationmeaslesofchildrenage

prevalenceofhivtotalofpopulation urbanpopulationgrowthannualspurb

improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year immunizationmeaslesofchildrenage

c.gdpgrowth##c.prevalenceofhivtotalofpopulation urbanpopulationgrowthannualspurb

improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

c.gdpgrowth##c.urbanpopulationgrowthannualspurb improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

urbanpopulationgrowthannualspurb c.gdpgrowth##c.improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

log close

16

GDP GROWTH ON MORTALITY RATE IN

2

000 AND 20

1

4

ACROSS THE WORLD

Yuxuan Tang

Jianing Wang

EC204 Empirical Economics II, Fall 2021

ABSTRACT

Journal of Population of Economics stated “For the period 1

8

00–2000, an increase in

GDP by 1% decreased mortality by 0.

7

%. This overall relationship is due to a strong

counter-cyclical relationship in the nineteenth century, which disappeared in the twentieth

century” (Svensson, M., Krüger, 20

10

). Based on the WorldBankData2years panel data in the

year of 2000 and 20

14

, this research mainly focused on the effects of GDP growth on mortality

rate, with different variables involved. The results showed a statistically insignificant relationship

between GDP growth and mortality rate. And 4

3

.3

9

% of the variation of the mortality rate can be

explained by GDP growth within the country when holding country level and time fixed effect.

1

https://learn.bu.edu/bbcswebdav/pid-9

5

75

6

30-dt-content-rid-590

11

850_1/xid-59011850_1

I. INTRODUCTION

This research aims to discover if GDP per capita affects mortality rate. Past research

shows that GDP per capita is inversely related to mortality rate during 1901-2000 in the United

States (M Harvey Brenner, 2005). In this article, Thomas McKeown demonstrated that

economic development is of fundamental importance to the decline of classic infectious and

childhood disease. With rapid economic growth in the 20th century, more people tend to have

vaccinations and are less vulnerable to infectious and childhood disease, which leads to a

decline in mortality rate. As a result, an inverse relationship between GDP per capita and

mortality rate worldwide was expected at the beginning.

After the hypothesis was conducted, we described and utilized a panel data across the

world in 2000 and 2014, and regressed GDP per capita and mortality rate with some variables

including improved sanitation facilities of po, co2 emissions metric tons per capita, improved

water, urban population growth annual spurb, prevalence of hiv total of population,

immunization measles of children age and others are tested with GDP per capita to find out how

it affects mortality rate. Then, we compiled our findings and found there is a statistically

insignificant relationship between the two main variables. Therefore, we use interactive variables

to test if the effects of GDP growth per capita on mortality rate depends on other variables listed

above. Then we created a graph that involves a linear regression and scatter plot were used to

make further comparison of fitness. Also, with the quadratic model being graphed, the turning

point is at 0.

16

4310932, and after this turning point, the relationship between GDP growth and

mortality rate becomes positive contrary to our expectations.

2

2. Literature Review

Many researchers had done studies relative to the effects of GDP growth on mortality rate

for years, and the reasons could be complicated. Mikael Svensson and Niclas A. Kruger used

wavelet methods to analyze the relationship between mortality rate and economic growth from

1800 to 2000 in Sweden. (Mikael Svensson and Niclas A. Krüger, 20

12

) According to the article,

it was found that in the early period of the 19th century, people were more vulnerable to disease

and health problems when the economy went downward. As a result, the mortality rate was

higher when the economy was poor. However, when we entered the 20th century, the augment

changed. People were more likely to stress out due to reasons including work stress, family

pressure due to unemployment, which leads to higher death rate. Furthermore, the research found

out some more specific factors that associate mortality rate with GDP growth, including stroke,

accident, suicide, cancer, and infection.

More findings were found by M Harvey Brenner. Using the time series model, with

variables of “ long-term effects of economic growth over 0–11 years,” “long-term effects of

unemployment over 0–11 years,” and “interactive effect of unemployment and GDP per capita

over 0–11 years”, it was found out that for a short period, increased mortality rate was due to

higher GDP growth, because of better technology with longer working period and speed.

However, for a longer period, GDP growth leads to the decline of mortality rate.（M Harvey

Brenner, 2005） More evidence was found by Brenner and Haines to prove this theory. According

to the article written by Haines in 2003, it was found that the United States experienced a rapid

economic growth but rising mortality rate between 1830 and 1860 due to deterioration of the

biological standard of living (Hanis 2003). During this period, the fast urban growth, mass

migration from abroad, changes in transportation infrastructure, rapid commercialization,

3

worsened the mortality environment which caused the mortality rate to rise. For a longer period,

Varvarigos constructed a model of a growing economy with pollution and testified that economic

growth and mortality rates are negatively related due to the difference of environment-related

structural parameters, such as lower p (units of pollution per output generating), which improves

the environmental conditions and reduces mortality rate (Varvarigos, 20

13

).

3. Data Description

Table 1

This research used panel data at country level worldwidely in the year of 2000 and 2014

from world bank data to analyze the relationship between GDP growth and mortality rate. A total

of 369 observations are collected from world bank data with 6 variables, including sanitation

facilities of po, co2 emissions metric tons per capita, improved water, urban population growth

annual spurb, prevalence of hiv total of population, and immunization measles of children age.

These 6 variables, together the two main variables are tested to find out the relationship between

GDP growth and mortality rate. The six variables are chosen because we realized that the higher

GDP a country has, the more conscious people have of their health. And as a result, more people

4

are getting vaccinated and actions or policies are taken for the sake of citizens’ health, which

leads to the decline of mortality rate.

The data of this research all come from world bank data, and two tables were created by

different years to describe the mean and standard deviation of the variables. Out of all the

variables, improved water has the highest mean value of 83.2% and 89.0 % in 2000 and 2014,

whereas urban population growth annual spurb have the lowest mean values around 2% in both

years.

Table 2

5

4. Model:

After we collected the data, we constructed a model of mortality rate as a function of GDP

growth at the country level of time fixed effect.

Within this fixed effect model, by holding year t and country i at constant level, mortrate

represents mortality rate, the continuous dependent variable in this equation, in year t and

country i. The main independent variable of this equation is gdpgrowth, which is continuous in

country i and year t, and is predicted to have a positive relationship with the main variable

mortality rate. The model is predicted as a linear regression as shown in the scatterplot graph. As

we used a time fixed effect model, the 6 other variables with i are absorbed into the ai variable

which change based on different countries. According to graphs shown below, most countries

with different mortality rates are scattered between 0% to 20% growth of GDP in both 2000 and

2014.

Also, in this model, the panel data at country level analyzes data from both year of 2000

and year of 2014 by using the dummy variable d00t and u is the error term. Graph1 represents

the worldwide GDP growth rate and mortality rate in 2000, and graph 2 displays GDP growth

rate and mortality rate in both the years of 2000 and 2014. However, by looking at the two

graphs below, we can see there is no inverse relationship between the GDP growth rate and

mortality rate, but instead a positive relationship. However, we cannot conclude that there is a

definitely positive relationship between GDP growth rate and mortality rate, as the dots mostly

concentrated in the middle of the graph rather than displaying a linear relationship. And there are

6

countries including Liberia, Equatorial Guinea , and Timor-Leste which are more than 3 standard

deviations away from the mean fall into the category of becoming outliers of the group.

Therefore, we used some interactive variables to test if there is a non linear relationship between

the two main variables. (shown in table-3)

Graph1: Scatterplot of Worldwide GDP Growth Rate and Mortality Rate in 2000.

7

Graph 2: Two-way Scatterplot of Worldwide GDP Growth and Mortality rate in 2000

and 2014.

8

5.RESULTS

Table 3: Regression Results

Looking at table 3, the coefficient of GDP growth rate has a statistically insignificant

relationship with mortality rate, and we can not conclude that GDP growth rate has a linear

relationship with mortality rate. Therefore, we added 6 more variables as shown in Table 3 that

are relative to mortality rate to test their relationships. The results in Model 2 show that the

coefficient of improved sanitation facility sanitation and CO2 emissions are statistically

insignificant with mortality rate. And the coefficient of Immunization measles of children’s age,

prevalence of HIV total of population, and improved water are statistically significant at 1%

level with mortality rate, with P-value equals to 0. Urban population growth annual spurb is

statistically significant at 5% level on the country level, with P-value equals to 0.047. Therefore,

we removed these two insignificant variables and ran the regression (Model 3). Since it’s a panel

data, in order to make sure different countries have the same coefficient effect, we uses country

9

level fixed effects, as we can see in Model 4, the coefficient of GDP growth rate still has a

statistically insignificant relationship with mortality rate, even after we controlling for the effects

of time (Model 5).

Furthermore with the data, we decided to add interactive variables of immunization and

GDP growth rate in Model 6, within in a country and after controlling for the effects of year,

with variable we testified significance before , the data shows that the coefficient of GDP growth

rate still has a statistically significant relationship with mortality rate at 1% level because the

effect of GDP growth on mortality rate depends on the percentage of Kids Immunization (12-13

months), and 77.5% of the variable of data in mortality rate explained by GDP growth rate

within country when the effects of time controlled.

In Model 7, we tested if the relationship of GDP growth rate to mortality rate depends on

other 3 significance variables. The results showed that the other 3 coefficients of interactive

variables are statistically insignificant which does not affect the relationship of GDP growth rate

to mortality rate. The results shown are not as consistent with our hypothesis, as immunization is

the factor that would affect the relationship between the two main variables.

6. Conclusions

Based on our findings on the model, GDP growth does not have a statistically significant

relationship with the mortality rate. However, when holding countries and time fixed, and we

added the interactive variable immunization, the results showed a statistically significant

relationship between GDP growth rate and mortality rate. As a result, we can conclude that GDP

growth rate and mortality rate depends on a third variable of immunization measles of children’s

age. And our findings do not completely support the hypothesis that there is a linear relationship

10

between GDP growth rate and mortality rate. As shown in graph 1 and graph 2, the countries are

scattered altogether in a group instead of displaying a linear relationship. While the independent

variable being statistically insignificant, the causes can be complicated due to many other

factors.

7. Limitations and future results

For further research in the future, we would apply more different variables which could

contribute to the change of the dependent variable, including import and export on the country

level. Also, besides testing the interactive variables, maybe we would use the log and square of

the independent variable after we collected the data. And more research will be done not only on

a country level, but within a country as well. There are some limitations in this research,

including not enough variables that are relative to the dependent variable. More variables will be

applied in the future research.

11

REFERENCES CITED

● Svensson, M., Krüger, N.A. Mortality and economic fluctuations. J Popul Econ 25,

12

15

–1235 (2012). https://doi.org/10.1007/s00148-010-0342-8

● M Harvey Brenner, “Commentary: Economic growth is the basis of mortality rate decline

in the 20th century—experience of the United States 1901–2000”, International Journal

of Epidemiology, Volume 34, Issue 6, December 2005, Pages 1214–1221. Retrieved

from https://doi.org/10.1093/ije/dyi146

● Svensson, Mikael, and Niclas A. Krüger. “Mortality and Economic Fluctuations:

Evidence from Wavelet Analysis for Sweden 1800—2000.” Journal of Population

Economics, vol. 25, no. 4, Springer, 2012, pp. 1215–35,

http://www.jstor.org/stable/23354789.

● M Harvey Brenner, Commentary: Economic growth is the basis of mortality rate decline

in the 20th century—experience of the United States 1901–2000, International Journal of

Epidemiology, Volume 34, Issue 6, December 2005, Pages 1214–1221,

https://doi-org.ezproxy.bu.edu/10.1093/ije/dyi146

● Varvarigos. (2013). ENVIRONMENTAL DYNAMICS AND THE LINKS BETWEEN

GROWTH, VOLATILITY AND MORTALITY. Bulletin of Economic Research, 65(4),

314–331. https://doi.org/10.1111/j.1467-8586.2011.00410.x

● Haines, Craig, L. A., & Weiss, T. (2003). The Short and the Dead: Nutrition, Mortality,

and the “Antebellum Puzzle” in the United States. The Journal of Economic History,

63(2), 382–413. https://doi.org/10.1017/S0022050703001839

12

https://doi.org/10.1007/s00148-010-0342-8

https://doi.org/10.1093/ije/dyi146

http://www.jstor.org/stable/23354789

https://doi-org.ezproxy.bu.edu/10.1093/ije/dyi146

https://doi.org/10.1111/j.1467-8586.2011.00410.x

https://doi.org/10.1017/S0022050703001839

APPENDIX A

DO FILE

clear all

set more off

capture

log close

cd/Users/sarah

use “/Users/sarah/Downloads/WorldBankData2years (7).dta”

sum mortrate gdpgrowth immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

improvedsanitationfacilitiesofpo

co2emissionsmetrictonspercapitae

urbanpopulationgrowthannualspurb improvedwater

*Graph 2

bysort year: eststo: estpost sum mortrate gdpgrowth

immunizationmeaslesofchildrenage

prevalenceofhivtotalofpopulation improvedsanitationfacilitiesofpo

co2emissionsmetrictonspercapitae urbanpopulationgrowthannualspurb improvedwater

esttab using summary_stats_table.rtf, cells((mean(fmt(%10.2f)) sd(fmt(%10.2f)))) label

title(Summary Statistics) nonumber nomtitle replace

label var mortrate “Mortality Rate”

label var gdpgrowth “GDP Growth”

label var immunizationmeaslesofchildrenage “% of Kids Immunization (12-13 months)”

label var prevalenceofhivtotalofpopulation “HIV population”

label var improvedsanitationfacilitiesofpo “improved sanitation facility”

label var co2emissionsmetrictonspercapitae “CO2 emissions”

label var urbanpopulationgrowthannualspurb “Urban Population Growth”

13

label var improvedwater “Imporve Water”

#delimit

;

esttab using SummaryStats1 , main(mean) aux(sd)

nonotes rtf replace label varwidth(30) modelwidth(9) b(%9.2f) nonumbers

mtitle(“2000” “2014”)

title(“Table 1: Summary Statistics by Year”)

addnotes(“NOTE: Table reports the mean and standard deviation in 2000 and 2014. ‘The

mean’ is above ‘standard deviation’ for each variable”)

;

#delimit cr

*Graphing in 2000 and 2014

twoway (lfitci mortrate gdpgrowth )(scatter mortrate gdpgrowth), ytitle(Mortality Rate)

*title(Two-way Scatterplot of GDP Growth Rate and Mortality Rate)

*graph save “Graph” “/Users/sarah/Desktop/Two-way Scatterplot 2000 and 2014.gph”

*Graphing in 2000

twoway (scatter mortrate gdpgrowth if year == 2000, mlabel(countryname)) (lfit mortrate

gdpgrowth)

*ytitle(Mortality Rate) xtitle(GDP Growth)

*title(Two-way Scatterplot of GDP Growth and Mortality Rate)

*graph save “Graph” “/Users/sarah/Desktop/Scatterplot 2000.gph”

*Graph 3

*regress X and Y

reg mortrate gdpgrowth, r

14

outreg2 using ResearchRegression , replace label title (“Regression Results”) adjr2 addtext

(Country FE, NO, YEAR FE, NO)

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, NO, YEAR FE, No)

*regression on full model

reg mortrate gdpgrowth immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

improvedsanitationfacilitiesofpo co2emissionsmetrictonspercapitae

urbanpopulationgrowthannualspurb improvedwater,r

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, NO, YEAR FE, No)

*running an F test on the insignificant variables improvedsanitationfacilitiesofpo

co2emissionsmetrictonspercapitae

test gdpgrowth improvedsanitationfacilitiesofpo co2emissionsmetrictonspercapitae

*The F test shows these variables are jointly insignificance, so we kick these variable out

*To see if gdpgrowth is jointly significance with other variable, let’s pick

immunizationmeaslesofchildrenage

test gdpgrowth immunizationmeaslesofchildrenage

*regression on full model

reg mortrate gdpgrowth immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

urbanpopulationgrowthannualspurb improvedwater,r

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, NO, YEAR FE, No)

*Regression on Fixed Effect

xtset countrynum year

xtreg mortrate gdpgrowth, r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, No)

15

xtreg mortrate i.year gdpgrowth, r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year c.gdpgrowth##c.immunizationmeaslesofchildrenage

prevalenceofhivtotalofpopulation urbanpopulationgrowthannualspurb

improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year immunizationmeaslesofchildrenage

c.gdpgrowth##c.prevalenceofhivtotalofpopulation urbanpopulationgrowthannualspurb

improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

c.gdpgrowth##c.urbanpopulationgrowthannualspurb improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

xtreg mortrate i.year immunizationmeaslesofchildrenage prevalenceofhivtotalofpopulation

urbanpopulationgrowthannualspurb c.gdpgrowth##c.improvedwater,r fe

outreg2 using ResearchRegressionTable , adjr2 addtext (Country FE, YES, YEAR FE, YES)

log close

16

Next Steps:

Now that you have submitted your data and have gotten feedback, here are some next steps for you to

work on to advance your paper.

1. Compile the articles that you will reference in your paper.

2. Read them.*

3. Write a 1-2 paragraph summary of each paper.**

4. Construct the citation of the papers for the References Cited section.

5. Begin writing your “Data Description” section.

6. Generate the Summary Statistics table.

7. Graph your variables.

8. Explore the correlation matrix.

9. Run regressions.

10. Generate your Regression Output table.

11. Decide on your overall (main) conclusion that your research has helped you discover.

12. Write your Results section.

13. Write your Summary/Conclusion section.

14. Write your Limitations and Future Research section.

15. Write your Literature Review section.***

16. Write your Introduction section.

17. Write your Abstract.

18. Check the Grading Rubric.

19. Ensure your format is correct, and that proper information is on your front page.****

*Not super intensely, just making sure you get the gist of what can be learned from the research.

**You will not use this entire summary in your paper, it just helps to write it RIGHT after reading it. You

will extract main findings later.

***Pay attention to how you cite a paper within text.

**** You must include your name(s), group number, “Spring 2020 EC204 [3:30 or 5pm] Section”, the

date and the abstract on the front page.

DETAILED VERSION ON FOLLOWING PAGE→

Next Steps (Detailed):

1. Compile the articles that you will reference in your paper.

a. Look for any that seem relevant to your research topic.

i. They may be about pre-discovered relationships with your DV or IV, or both.

ii. They don’t need to address your exact research question. Instead, they may

have discovered something about relationships that include either of your main

variables. However, they need to “fit” into your overall story/analyses and

should not seem completely unrelated.

2. Read them.

a. You just want to get the gist of what can be learned from the research.

b. If the econometrics is beyond the scope of our course (Metrics 1!), then just make sure

you can understand their main conclusions.

3. Write a 1-2 paragraph summary of each paper.

a. Make sure YOU ARE NOT copying/pasting from the abstract, or from anywhere else.

4. Construct the citation of the paper for the References Cited section.

a. APA style

i. By author:

1. https://owl.purdue.edu/owl/research_and_citation/apa_style/apa_for

matting_and_style_guide/reference_list_author_authors.html

ii. More detailed info on references to articles in particular:

1. https://owl.purdue.edu/owl/research_and_citation/apa_style/apa_for

matting_and_style_guide/reference_list_articles_in_periodicals.html

b. MLA style is also fine if that’s the one on which you are already well-versed (just be

consistent within your References Cited section).

5. Begin writing your “Data Description” section.

a. This should be straight to the point:

i. Source(s) of data

ii. Describe your data.

1. Who or what make up your elements/observations?

2. Describe your variables:

a. Include tables: Variables’ Descriptions and Summary Statistics.

i. You can copy/paste the one you have submitted, with

any changes I recommended in feedback.

b. Don’t include polynomial or log terms in your description, just

describe the underlying X

b. Example: “Data were obtained from the World Bank and span 1980-2015. The list of

countries is provided below the Summary Statistics Table (Table 2) and the variables are

described in Table 1.”

i. Of course include anything else about your data that you think may be

important for the reader to know, including missing countries for example, or

which “population of interest” you think the sample was drawn from.

c. Note: this is NOT the place to mention potential sources of omitted variable bias.

CONTINUES ON FOLLOWING PAGE→

https://owl.purdue.edu/owl/research_and_citation/apa_style/apa_formatting_and_style_guide/reference_list_author_authors.html

https://owl.purdue.edu/owl/research_and_citation/apa_style/apa_formatting_and_style_guide/reference_list_author_authors.html

https://owl.purdue.edu/owl/research_and_citation/apa_style/apa_formatting_and_style_guide/reference_list_articles_in_periodicals.html

https://owl.purdue.edu/owl/research_and_citation/apa_style/apa_formatting_and_style_guide/reference_list_articles_in_periodicals.html

6. Generate the Summary Statistics table.

a. Use the resources on Blackboard. Find the do file and example of a summary statistics

table under “Research Project Stuff→Stata Stuff→Stata Help Videos, Commands,

etc.→Producing Publication-Style Tables in Stata→Summary Statistics Tables”

b. Make sure you use the label option in the esttab command to ensure that Stata uses the

labels of the variables and not the (often ugly or uninformative) variable names

themselves.

7. Graph your variables.

a. Scatterplots, bar charts, pie charts… whatever help you better understand your data and

whatever may help your reader either better understand or become more motivated by

your research.

b. This is where you can set yourself apart from the rest of the class, by providing a graph

that is super informative and more than a simple basic scatterplot.

8. Explore the correlation matrix.

a. Use the corr command to see the correlation matrix.

b. This may help you describe omitted variable bias that is removed once the variable is no

longer omitted, or also help you discover other possible control variables or other

interesting relationships!

i. The correlation matrix does NOT need to be reported in your paper unless you

think it should be/is extremely helpful to make a point you are trying to make.

9. Run regressions.

a. Play with various model specifications: logs, polynomials, interaction terms…

10. Generate your Regression Output table.

a. Use outreg2.

i. Find Resources on Blackboard under “Research Project Stuff→Stata Stuff→Stata

Help Videos, Commands, etc.→Producing Publication-Style Tables in

Stata→Regression Results Tables

b. The first column will be just your DV regressed on your main variable of interest

i. No controls, no “fixed effects” (for panel data)

c. The subsequent columns will display the various models you attempted

i. Don’t add in controls one at a time

ii. Do add in time and entity fixed effects one at a time (if panel data)

d. Be sure your table reports the adjusted R2 for each model, as well as the number of

observations.

e. You should only have ONE regression output table (MAYBE two if you run LOTS of

interesting regressions)

i. Each different regression is a different column, not a different table

ii. You can often fit up to 5-7 columns in one table

iii. Make sure that your columns clearly indicate what your DV is.

11. Decide on your overall (main) conclusion that your research has helped you discover.

a. What’s the main finding? It may or may not be exactly what you set out to discover.

b. This will shape the way you “frame” your research paper, keeping a consistent flow

between sections/ideas that have an overall (and consistent) point throughout, all

leading to this overall conclusion.

CONTINUES ON FOLLOWING PAGE→

12. Write your Results section.

a. This should also be consistent with the “frame” of the paper.

b. It describes things you can learn from the regression output table.

13. Write your Summary/Conclusion section.

a. This should also be consistent with the “frame” of the paper.

b. Unlike your Results section, it only discusses the main takeaway.

c. This is where any policy implications would be discussed. How does what we have

learned from your research help us better understand what we can do to achieve a

particular goal?

i. Note that some papers may not have “policy implications” but instead may

prescribe a perspective we should share, a new way to look at something, or

provide advice on how to approach future decision making.

14. Write your Limitations and Future Research section.

a. This is where you discuss the limitations of your research, including missing data that

may be biasing your results

b. You also mention a possible direction (or two, or three) that research can take.

15. Write your Literature Review section.***

a. Be sure your lit review section is very concise, only mentioning the main takeaways from

papers that help us better understand something about either your DV or your main IV

(or both).

b. Be sure you think about how you are framing your research paper to make sure these

additions fit nicely and flow smoothly toward your own main conclusions.

16. Write your Introduction section.

a. This is generally saved for last because by now you know EXACTLY how you are framing

your research conclusions.

b. This section contains a (brief) description of your research question, the motivation

(why is it worth studying), and your main conclusions.

17. Write your Abstract.

a. A VERY brief description of your research question and your main conclusions.

18. Check the Grading Rubric.

a. This will help ensure you hit all/most of the necessary pieces.

19. Ensure your format is correct, and that proper information is on your front page.****

**** You must include your name(s), group number, “Spring 2020 EC204 [3:30 or 5pm ]Section”, the

date and the abstract on the front page.

The price is based on these factors:

Academic level

Number of pages

Urgency

Basic features

- Free title page and bibliography
- Unlimited revisions
- Plagiarism-free guarantee
- Money-back guarantee
- 24/7 support

On-demand options

- Writer’s samples
- Part-by-part delivery
- Overnight delivery
- Copies of used sources
- Expert Proofreading

Paper format

- 275 words per page
- 12 pt Arial/Times New Roman
- Double line spacing
- Any citation style (APA, MLA, Chicago/Turabian, Harvard)

Delivering a high-quality product at a reasonable price is not enough anymore.

That’s why we have developed 5 beneficial guarantees that will make your experience with our service enjoyable, easy, and safe.

You have to be 100% sure of the quality of your product to give a money-back guarantee. This describes us perfectly. Make sure that this guarantee is totally transparent.

Read moreEach paper is composed from scratch, according to your instructions. It is then checked by our plagiarism-detection software. There is no gap where plagiarism could squeeze in.

Read moreThanks to our free revisions, there is no way for you to be unsatisfied. We will work on your paper until you are completely happy with the result.

Read moreYour email is safe, as we store it according to international data protection rules. Your bank details are secure, as we use only reliable payment systems.

Read moreBy sending us your money, you buy the service we provide. Check out our terms and conditions if you prefer business talks to be laid out in official language.

Read more