tiểu luận kinh tế lượng tài chính factors affecting the human development index over the world in 2017

Due to the limited of data resources, we can only pick up a few prominent factors of those countries in 2017, which are Life expectancy at birth LE, Expected years of schooling EYS - mea

FOREIGN TRADE UNIVERSITY FACULTY OF INTERNATIONAL ECONOMICS



ECONOMETRICS REPORT Factors affecting the Human Development Index over the world in 2017

Group: 1 Trinh Thi Hong Nhung – ID: 1813340049

2 Nguyen Hoang Ha – ID: 1813340023

3 Vu Phuong Linh – ID: 1813340036

4 Pham My Duyen – ID: 1813340019

5 Nguyen Binh An – ID: 1713340002

6 Dao Minh Hoa – ID: 1212340030 Class: KTEE301.1 (21/10-15/12/2019)

Instructors: Dr Nguyen Thuy Quynh

Hanoi, December 2019

FOREIGN TRADE UNIVERSITY FACULTY OF INTERNATIONAL ECONOMICS



ECONOMETRICS REPORT

Factors affecting the Human Development Index over the world in 2017

Group 1 Trinh Thi Hong Nhung – ID: 1813340049

2 Nguyen Hoang Ha – ID: 1813340023

3 Vu Phuong Linh – ID: 1813340036

4 Pham My Duyen – ID: 1813340019

5 Nguyen Binh An – ID: 1713340002

6 Dao Minh Hoa – ID: 1212340030

Class: KTEE301.1 (21/10-15/12/2019)

Instructors: Dr Nguyen Thuy Quynh

Hanoi, December 2019

TABLE OF CONTENT

ABSTRACT 4

INTRODUCTION 5

SECTION 1 OVERVIEW OF THE TOPIC 7

1 The Human Development Index (HDI) 7

1.1 HDI stands for Human Development Index 7

1.1.1 Forming indices for each of the four metrics 7

1.1.2 Aggregating the four metrics to produce the HDI 7

1.2 Economic theories 8

1.2.1 The effect of Life expectancy at birth on HDI 8

1.2.2 The effect of Expected years of schooling on HDI 8

1.2.3 The effect of Gross national income (GNI) per capita on HDI 9

1.2.4 The effect of Fertility on HDI 9

1.2.5 The effect of Inflation on HDI 10

1.3 Former researches 10

SECTION 2 Model Specification 11

1 Methodology in the study 11

1.1 Method to derive the model 11

1.2 Method to collect and analyze the data 11

1.2.1 Collect the data: 11

1.2.2 Analyze the data: 11

1.3 Theoretical model specification 11

1.3.1 Specification of the model: 11

1.3.1.1 Population Regression Model: 12

1.3.1.2 Sample Regression Model 12

1.3.2 Explanation of the variables: 12

1.4 Description of the data 13

1.4.1 Statistical description of the variables 13

1.4.2 Correlation matrix between variables 14

SECTION 3 ESTIMATED MODEL AND STATISTICAL INFERENCE 16

1 Estimated Model 16

1.1 Estimation result 16

1.2 The sample regression model 16

1.3 The coefficient of determination 17

2 Meanings of estimated coefficients 17

3 Other results analysis 18

4 Test for model’s possible problems and correct them: 19

4.1 Testing omitted variables (verifying the model's correct format) 19

4.2 Testing multi-collinearity phenomenon: 19

4.3 Testing for Heteroskedasticity 20

4.4 Self-correlation test 20

4.5 Test the normal distribution of random errors 20

5 Hypothesis Testing 21

5.1 Testing the significance of an individual regression coefficient β j 21

5.1.1 The confidence interval approach 21

5.1.2 The T-distribution approach 22

5.1.3 The P-value approach 23

5.2 Testing the significance of the model 23

5.2.1 The P-value approach 23

6 Recommendations 25

6.1 Recommendations to increase the expected years of schooling and improve the education system 25

6.2 Recommendations to increase the life expectancy 26

7 Recommendations to decrease inflation rate 27

CONCLUSION 28

REFERENCES 29

APPENDIX 30

INDIVIDUAL ASSESSMENT 35

We are living in a complex world People, nations and economies are more connectedthan ever, and so are the global development issues we are facing These issues spanborders, straddle social, economic and environmental realms, and can be persisting orrecurring From urbanization to the creation of jobs for millions of people, the world’schallenges will only be solved using approaches that take both complexity and localcontext into account For almost thirty years, UNDP’s human development approach—with its emphasis on enlarging people’s freedoms and opportunities rather than economicgrowth—has inspired and informed solutions and policies across the world

In hope of providing a deeper insight, scrutinizing a specific case, our group would like to

take the topic “Factors affecting the Human Development Index over the world in

2017” in thorough consideration This report investigates the determinants of Human

Development Index in 120 countries, employing the methods of panel data analysis Due

to the limited of data resources, we can only pick up a few prominent factors of those

countries in 2017, which are Life expectancy at birth (LE), Expected years of

schooling (EYS) - mean years of school, Gross national income (GNI) per capita, Inflation rate (INF) and Fertility rate (FER) Our research indicates that the relationship

between human development index and three factors which are GNI per capita, inflationrate (INF), fertility rate (FER) is negative A reverse tendency could be observed in therelationship between HDI and life expectancy at birth (LE), expected years of school(EYS), as they are positive

Econometrics is the quantitative application of statistical and mathematical models usingdata to develop theories or test existing hypothesis in economics and to forecast futuretrends from historical data It subjects real-world data to statistical trials and thencompares and contrasts the results against the theory or theories being tested

Depending on whether you are interested in testing an existing theory or in using existingdata to develop a new hypothesis based on those observations, econometrics can besubdivided into two major categories: theoretical and applied Those who routinelyengage in this practice are commonly known as econometricians

In the report, we will use the econometric model to find out the relationship between byusing collected data from World Bank, UNDP and others sources, whether they havepositive or negative relationship And from the result, we may have somerecommendations to countries with lower human development index

We recognize the important of econometrics in social economics In order to understandhow the econometrics works in real life and to apply econometrics effectively andcorrectly, our group would like to develop a report under the guidance of Dr NguyenThuy Quynh In this report, we used the econometrics analysis tool Stata to analyze thetopic “Factors affecting the Human Development Index over the world in 2017”

The report contains the following contents:

 SECTION 1: OVERVIEW OF THE TOPIC

 SECTION 2: MODEL SPECIFICATION

 SECTION 3: ESTIMATED MODEL AND STATISTICAL INFERENCES

may hardly avoid mistakes We are always willing to receive your comments so that ourgroup can improve and complete this report.

Many thanks!

SECTION 1 OVERVIEW OF THE TOPIC

1 The Human Development Index (HDI)

The Human Development Index (HDI) is a composite statistic of life expectancy,

education, and per capita income indicators, which are used to rank countries into four tiers of human development A country scores higher HDI when the lifespan is higher, the education level is higher, the GDP per capita is higher, the fertility rate is lower, and the inflation rate is lower

1.1 HDI stands for Human Development Index

It was developed and launched by Pakistani economist Mahbub-ul-Haq, followed byAmartya Sen, an Indian economist, in 1990 Human Development Index, HDI, is acomprehensive tool devised by the United Nations for measuring the levels of social andeconomic developments of the different countries and ranking them accordingly It is acomparative measure of life expectancy, education, literacy, and standard of living.Essentially, Human Development Index, HDI, makes use of four parameters formeasuring and ranking countries according to their social and economic developmentwhich includes the Life Expectancy at Birth, Expected Years of Schooling, Mean Years of

There are two steps to calculating the HDI:

1.1.1 Forming indices for each of the four metrics

The values of each of the four metrics are first normalized to an index value of 0 to 1 To

do this, “goalposts” of the maximum and minimum limits on each metrics are set by theUNDP, as shown in the table With the actual value for a given country, and the globalmaximum and minimum, the dimension (indices) value for each metric is calculated as:

The dimension index is therefore 1 in a country that achieves the maximum value and it is

0 for a country that is at the minimum value

1.1.2 Aggregating the four metrics to produce the HDI

Once each of the individual indices has been calculated, they are aggregated to calculate

the HDI The HDI is calculated as the geometric mean (equally-weighted) of lifeexpectancy, education, and GNI per capita, as follows:

The education dimension is the arithmetic mean of the two education indices (mean years

of schooling and expected years of schooling)

1.2 Economic theories

The main purpose of our group’s research is to determine the factors which affect thefluctuation of The Human Development Index (HDI) However, we will mainly focus onthe long-term relationship between those factors and HDI According to previouslypublished researches, some long-term factors substantially affecting HDI are Lifeexpectancy at birth, expected years of schooling, Gross national income (GNI) per capita,Inflation, and Fertility

1.2.1 The effect of Life expectancy at birth on HDI

Life expectancy at birth (years) is the average number of years a newborn child would live

if current mortality patterns were to stay the same.According to Max Roser in an article on “Our World in Data” website, the firstcomponent of the HDI – a long and healthy life – is measured by life expectancy Long-run estimates of life expectancy across the world are shown in the visualization Forcountries where historical records are available, such as the UK, estimates can extend asfar back as 1543 – click on the UK to see this long-run perspective Global and regionalestimates extend back to the year 1770 This dataset is based on a combination of datafrom the Clio Infra project, the UN Population Division, and global and estimates forworld regions from James Riley (2005)

1.2.2 The effect of Expected years of schooling on HDI

The second component – access to education – is measured by expected years of schooling of children at school-entry age and mean years of schooling of the adult

population

Education has been one of the most integral drivers and outcomes of global development.The provision of education is now viewed in most parts of the world as a basic right –with pressure on governments to ensure a high-quality education for all

Education should have a positive effect on HDI because as education increases so does theknowledge of how to lead a healthier life This knowledge might, for example, take theform of improved nutrition or reduced exposure to various health risks, such as indoorpollution exposures Education is measured by the education index In this analysis, theTotal literacy rate has been taken as a proxy of Education index.There are many metrics we can use to assess education access, quality, and attainment –

we cover many of them throughout our work on educationThe visualizations present the two metrics that the HDI captures:

- Mean years of schooling estimates the average number of years of total schooling adults

aged 25 years and older have received This data extends back to the year 1870 and isbased on the combination of data from Lee and Lee (2016); Barro-Lee (2018); and the UN

- Expected years of schooling measures the number of years of schooling that a child of

school entrance age can expect to receive if the current age-specific enrollment ratespersist throughout the child’s life by country

1.2.3 The effect of Gross national income (GNI) per capita on HDI

The architects of the HDI have decided to add a third dimension – a decent standard ofliving – and to measure it by Gross National Income per capita.GNI is expected to be positively related to HDI for diverse reasons GNI displaysdisposable income As disposable income increases, people have more resources for bettershelter, food, and medical care Again, countrywide data might offer some advantagesover individual data: a wealthy person living in a poor country is unlikely to have thesame access to quality food and medicine as a wealthy person living in a wealthy country.Since income is highly correlated with many other categories that would affect HDI (e.g.education, life expectancy), it is also held constant to estimate, without bias, the specific

For most of human history, our ancestors were stuck in a world of poor health, hunger andlittle access to formal education Economic growth – particularly over the past fewcenturies – has allowed some part of the world population to break out of these conditions.The map shows the GNI per capita - this is the metric that the HDI relies on:

1.2.4 The effect of Fertility on HDI

Total Fertility rate (Children per woman): The number of children that would be born toeach woman with prevailing age-specific fertility rates

where ASBR is each five-year age-specific birth rate defined as

where Bx is the number of live births to mothers age x and Px is the number of resident

The higher the fertility is, the higher the population expected can get It might result in ashortage of sources of food and drinks, occupations, education services, accommodations,etc, which decreases HDI

1.2.5 The effect of Inflation on HDI

Inflation (annual %): Inflation, as measured by the annual growth rate of the GDP implicitdeflator, shows the rate of price change in the economy as a whole The GDP implicitdeflator is the ratio of GDP in current local currency to GDP in constant local currency

The inflation rate is the rate of increase in the price level of the economy It shows thelevel of inflation of the economy Normally, the inflation rate is calculated based on the

consumer price index or the GDP deflator.The rising in goods and services leads to the rising in inflation Inflation has a negativerelationship with GDP deflator Whereas GDP is one of the basic indicators to evaluateeconomic development If a country's GDP declines, the economy will decline and alsoHDI.

1.3 Former researches

Almost former researches I have ever read can mention the right definition of HDI and the

3 most important factors affecting it: life expectancy, education, and living standard.Some of them can point out many detailed factors with data tables, which express therelationship between HDI and factors However, the authors might not extremely investmuch time and energy in explaining in words and proving theoretically for the way factorsaffect HDI fluctuation Their main factors are also different from ours For instance, in the

“ Determinants of Human Development Index: A cross- Country Empirical”, Smit Shadhas written 6 factors, but he only explains two of them and just give the definitions forothers There is no evidence showing how life expectancy, education, income, GNA,inflation, fertility, etc affect HDI as well as social, economic problems

SECTION 2 MODEL SPECIFICATION

1 Methodology in the study

1.1 Method to derive the model

The process in this research is called Multiple Linear Regression This is a linear approach

to modeling the statistical relationship of a dependent variable on one or more explanatoryvariables Specificially, in our research, it is the statistically dependent relationship of

Human Development Index on Life Expectancy, the Expected years of schooling, Gross National Income per capita, Inflation rate and Fertility rate.

1.2 Method to collect and analyze the data

1.2.1 Collect the data:

Collected data are secondary data, in form of Data, showing the numerical information ofsome factors of over 120 countries all over the world in 2017 The data were collectedfrom World Bank and UN which has a very high level of precision

1.2.2 Analyze the data:

Our group has used Stata to analyze the dataset and interpret the correlation matrixbetween variables

1.3 Theoretical model specification

1.3.1 Specification of the model:

According to previous published researches, our group has established a function toanalyze the relationships between the above factors and the human development index aswell as the effects of those variables toward the dependent variable:

HDI = f(LE, EYS, GNI, INF, FER)

Where:

+ LE: Life Expectancy at birth (years)

+ EYS: Expected years of schooling (years)

+ GNI: Gross National income per capita ($)

+ INF: Inflation rate (%)

+ FER: Fertility rate (%)

Thus, according to the economic theories, in order to analyze the factors influencing the

Human Development index, our group has discussed and decided to choose the

regression analysis models

1.3.1.1 Population Regression Model:

PRF: HDI =β1+β2¿it+β3ln EYS it+β4lnGNI it+β5INF it+β6FER it+u it

Where:

 β1: the intercept term of the model

 β2: the regression coefficient of “Life Expectancy” ¿it

 β3: the regression coefficient of “Expected years of schooling” lnEYS it

 β4: the regression coefficient of “Gross National Income per capita” lnGNI it

 β5: the regression coefficient of “Inflation rate” INF it

 β6: the regression coefficient of “Fertility rate” FER it

 u it: the disturbance term of the model, represents other factors that affect HDI

but not mentioned in the model

1.3.1.2 Sample Regression Model

SRF: HDI it= ^β1+ ^β2¿it+ ^β3ln EYS it+ ^β4ln GNI it+ ^β5INF it+ ^β6INF it+ ^u it

 u^it: the estimator of u it− ¿the residuals term

1.3.2 Explanation of the variables:

Table 1: Expected impact of explanatory variables on explained variable

Regression coefficient

3 lnEYS Expected years of schooling years + ¿

4 lnGNI Gross National Income per

capita

● The dependent variable is HDI.

● The explanatory variables are LE, lnEYS, lnGNI, INF and FER.

1.4 Description of the data

Data sources The dataset was collected from the official website of World Bank and UN,includes 120 observations of 120 countries all over the world in 2017

1.4.1 Statistical description of the variables

Run the command sum LE lnEYS lnGNI INF FER to interpret the dataset, the result

obtained including the number of observations (Obs), the average value (Mean), thestandard deviation (SD) as well as the minimum (Min) and maximum (Max) values ofeach variable as the table belows:

Table 2: Statistical description of the variables

No Variable Obs Mean SD Min Max

1.4.2 Correlation matrix between variables

Run the command corr LE lnEYS lnGNI INF FER to analyze the correlation between

the variables, we have the result is the table of correlation matrix between variables:

Table 3: Correlation matrix between variables

According to the Correlation matrix between variables:

 The correlation coefficient between LE and HDI is 0.9010, which is positive and so high Therefore, LE has a positive effect on HDI, any change in the life expectancy will lead to a covarieted change very much in the Human

Development index.

 The correlation coefficient between EYS and HDI is 0.9185 which is positive

and very high Therefore, EYS has a positive effect on HDI, any change in the

Expected years of schooling will just lead to a big change in the Human Development index.

 The correlation coefficient between GNI and HDI is 0.9608, which is positive

and extremely high Therefore, GNI has a positive effect on HDI, a slight change in the Gross national income (GNI) per capita will lead to a major inverted change in the Human Development index.

 The correlation coefficient between INF and HDI is −0.4890, which is negative

and quite high Therefore, INF has a negative effect on HDI, which means the

Inflation rate has a slight inverted change in the Human Development index.

 The correlation coefficient between FER and HDI is −0.8073, which is negative

and high Therefore, FER has a negative effect on HDI, which means the HDI has a huge influence on Human Development , any change in the Fertility rate will lead to a lot of inverted fluctuations in the Human Development index.

SECTION 3 ESTIMATED MODEL AND STATISTICAL INFERENCE

1 Estimated Model

1.1 Estimation result

Run the command reg HDI LE lnEYS lnGNI INF to compute the estimation result,

the result obtained is a table:

Table 4: Estimated Model

Estimated sum of square (ESS) 2.41301155

Residual sum of square (RSS) 0.040735121

Variables Coefficient ^β T P-value Confident interval (95%)

1.2 The sample regression model

We have the Sample Regression Model:

HDI it= ^β1+ ^β2¿it+ ^β3ln EYS it+ ^β4ln GNI it+ ^β5INF it+ ^β6FER it+ ^u it

According to the estimated result from Stata using the Ordinary Least Squares (OLS)

method, we obtained the Sample Regression Function (SRF) as below:

HDI it=−0.744174 +0.0045675 ¿it+0.1983033 lnEYSit+0.0673941lnGNI it−0.0002993 INFit−0.0019004 FER it+ ^u it

1.3 The coefficient of determination

The coefficient of determination R2 (R-squared) ¿ 0.9834 means 98.34% of the total

variation in the dependent variable, which is Human Development index, is explained by

the explanatory variables, which are Life expectancy at birth, Expected years of

schooling , Gross national income (GNI) per capita , Inflation rate, Fertility rate; the

remains are due to other factors

Besides the coefficient of determination, we also take the adjusted R2 (´R2) under

consideration, because adding more variables into the model can sometimes make the R2

less significant The ´R2 of the model is also very high: about 0.9827, not so different from

the R2, which mean the model can still explain for 98.27% of the fluctuations and the

variables added are reasonable

2 Meanings of estimated coefficients

 The constant term is estimated to be ^β1=−0.744174: Holding every explanatory

variable equals to 0, the expected value of Human Development index (HDI) will

be −0.744174

 The regression coefficient of LE is estimated to be ^β2=0.0045675: Holding every

explanatory variables unchanged, if the Life expectancy at birth (LE) increases by

1 year, the expected value of Human Development index (HDI) will increase by

0.1983033

 The regression coefficient of lnEYS is estimated to be ^β3=0.1983033: Holding other

explanatory variables unchanged, if the Expected years of schooling (EYS) increases by 1 year, the expected value of Human Development index (HDI) will

increase by 0.1983033

 The regression coefficient of lnGNI is estimated to be ^β4=0.0673941: Holding other

explanatory variables unchanged, if the Gross national income (GNI) per capita increases by 1$, the expected value of Human Development index (HDI) will

increase by 0.0673941

 The regression coefficient of INF is estimated to be ^β5=0.0002993, which is contrary

to expectation and correlation matrix Besides, its confident interval (95%) is

[−0.00114; 0.0017387] which included 0 Therefore, INF isn’t statistical significant.

 The regression coefficient of FER is estimated to be ^β6=−0.0019004: Holding other

explanatory variables unchanged, if the Fertility rate (FER) increases by 1%, the expected value of Human Development index (HDI) will decrease by 0.0019004.But its confident interval is [−0.0071364 ;0.0033356 ¿ which included 0 Therefore,

FER isn’t statistical significant.

3 Other results analysis

 Number of observations: Obs ¿ 120