tiểu luận kinh tế lượng ANALYSIS OF FACTORS AFFECTING FOREIGN DIRECT INVESTMENT (FDI) IN THE WORLD

Foreign direct investment FDI is recognized as a powerful engine for economic growth.. Literature Review - Foreign direct investment FDI: An investment made by a firm or individual in o

FOREIGN TRADE UNIVERSITY FACULTY OF BUSINESS ADMINISTRATION

- 

 -REPORT OF ECONOMETRICS

ANALYSIS OF FACTORS AFFECTING FOREIGN DIRECT INVESTMENT (FDI) IN THE WORLD

Class ID: KTEE309(1-1920).2_LT

Lecturer: PhD Dinh Thi Thanh Binh

Hanoi, 2019

TABLE OF CONTENTS

INTRODUCTION 3

CONTENT 4

I Literature Review and Statistic Description 4

1 Literature Review 4

2 Main content 5

II Statistic Description 5

1 VARIABLE DESCRIPTION 5

a Running DES function 5

b Running SUM function 6

c Running TAB1 function 6

2 REGRESSION AND CORRELATION 8

a Set up model 8

b Analysing the correlation between independent variables 9

c Correlation relationship between variables 9

d Running function 10

3 TESTING MODEL 11

a Multicollinearity 11

b Heteroskedasticity 13

c Testing Multiple Linear Regressions: F Test 14

d Testing hypothesis in statistic 15

IMPLICATION AND CONCLUSION 16

In a growing society, econometrics has been a science with many practical applications, especially issues related to human social life Econometrics provides powerful tools that enables economists to analyze the collected statistics and make predictions about social phenomena As students in the economic sector, we are well aware of the need to study and research econometrics

Foreign direct investment (FDI) is recognized as a powerful engine for economic growth It enables capital-poor countries to build up physical capital, create employment opportunities, develop productive capacity, enhance skills of local labor through transfer of technology and managerial know-how, and help integrate the domestic economy with the global economy

Therefore, the identification of factors affecting the attraction of foreign investment and an analysis of the impact of each factor to attract foreign investment is essential for the government in offering policies to attract investment capital And this aspect

has inspired our group to illustrate it detailedly with this report “Analysis of Factors Affects FDI in the world”.

In “Analysis of Factors Affecting FDI in the world”, factors will be given in details

and collected data of following factors will be operated, calculated and explained Further description will be continuously attached to each section for further understandings By the end of this report is followed by a reasonable conclusion section to summarize our result as a whole

We would like to thank our instructor - PhD Dinh Thi Thanh Binh for helping us with this paper In the process of making the report, although we have tried hard, but surely could not avoid the errors, we hope that you will contribute to the completion of our report!

I Literature Review and Statistic Description

1 Literature Review

- Foreign direct investment (FDI): An investment made by a firm or individual in one country into business interests located in another country in a year

- Population growth (pog): The rate of the increase in the number of individuals

in a population Large populations provide a large market for products and services, have a large labor force and a vast skill base Considering the advantages of a large population, it was hypothesized that investors would make larger investments in countries with larger populations

- Inflation rate (ir) :Inflation is a quantitative measure of the rate at which the

average price level of a basket of selected goods and services in an economy increases over a period of time Rate of inflation is a crucial factor in influencing the inflow of foreign investment A high rate of inflation signifies economic instability associated with inappropriate government policies

- Labor costs (lac) : The cost of labor is the sum of all wages paid to employees,

as well as the cost of employee benefits and payroll taxes paid by an employer Wage as an indicator of labour cost has been the most contentious of all the potential determinants of FDI

- Public debt (pud) : The public debt is how much a country owes to lenders

outside of itself These can include individuals, businesses, and even other governments The study employs a Vector Error Correction Model, which provides both the long run and short run relationships among the variables The long run results indicate that the relationship between public debt and foreign direct investment, as well as interest rate and FDI, is positive and statistically significant: the level of public debt should increase so that the level of foreign direct investment can increase in the country

2 Main content

a) Variables: There are 5 chosen variables:

 Fdi: Dependent variable

 Pog: Independent variable

 Ir: Independent variable

 Lac: Independent variable

 Pud: Independent variable

b) Outline: The report includes 4 main parts :

 Part 1 : Variables description using functions DES, TAB, SUM

 Part 2 : Analyze regression model and correlation

 Part 3 : Testing model

 Part 4 : Conclusion

II Statistic Description

1 VARIABLE DESCRIPTION

a Running DES function

The most important information after using DES function is the variables label

des fdi pog ir lac pud

Table 1 Result from running DES function

By using des, we know clearly about the variables According to the results, we know:

 fdi: Foreign direct investment in a year (unit: USD)

 pog: Population growth in a year (unit: percentage)

 ir: Inflation rate in a year (unit: percentage)

 lac: Labor costs in a year (unit: USD)

 pud: Public debt in a year (unit: USD)

b Running SUM function

SUM function lets us know about observations, mean, standard deviation, max and min value of the variables.

sum fdi pog ir lac pud

Table 2 Result from using SUM function

By using SUM function, we have:

 fdi: With 180 observations, the mean FDI per year is -1.84e+09, Std Dev is

2.59e+10 The minimum average FDI is -2.02e+11, the maximum average FDI

is 1.332+11

 pog: With 216 observations, the mean population growth per year is 1.303029,

Std Dev is 1.307686 The minimum population growth is -3.91335, the maximum population growth is 5.790631

 ir: With 203 observations, the mean inflation rate per year is 1.70359, Std.

Dev is 8.514521 The minimum inflation rate is -36.56478, the maximum inflation rate is 38.88166

 lac: With 192 observations, the mean Labor costs per year is 19352.66, Std.

Dev is 20301.35 The minimum Labor costs is 750, the maximum Labor costs

is 121090

 pud: With 122 observations, the mean Public debt per year is 5.29e+10, Std.

Dev is 1.52e+11 The minimum Public debt is 1.17e+08, the maximum Public debt is 1.33e+12

c Running TAB1 function

Using TAB1 function allows to describe more than 1 variables coincidently with

frequency and percent of the variables

tab1 fdi pog ir lac pud

Table 3 Result from running TAB1 with FDI in a year.

Example for analyzing information from the table:

FDI in a year ranges from -2.02e+11 to 1.33e+11.

2 REGRESSION AND CORRELATION

a Set up model

Regression displaying the relationship between the dependent variable Y – Foreign

Direct Investment (FDI), and independent variables pog (X 1 ), ir (X 2 ), lac (X 3 ), pud (X 4 ) has the following form:

General Regression ModelFDI=β0+β1× pog+ β2×ir +β3×lac+ β4× pud + ^u i

Sample Regression Model

^

FDI= ^β 0+ ^β 1 pog + ^β 2 ir + ^β 3 lac+ ^β 4 pud

b Analysing the correlation between independent variables

Using function:

corr fdi pog ir lac pud

The result is as below:

Table 4 Result from running CORR between independent variables

c Correlation relationship between variables

 FDI and pog is positive

The higher the population growth rate, the more FDI will be invested

 FDI and ir is positive

The higher the Inflation rate is, the more FDI will be received

 FDI and lac is negative

The higher the labour cost/worker wage is, the fewer FDI will be received

 FDI and pud is negative

The higher the public debt is, the fewer FDI will be invested

Types of correlation relationship:

 0.1 > r: no correlation

 0.1 < r < 0.3: weak correlation

 0.3 < r < 0.5: medium correlation

 0.5 < r: quite strong correlation

Overall, the independent variables do not have a quite strong correlation relationship

with the dependent one Especially, the variable “ir” has a very weak relationship.

d Running function

Using function:

reg fdi pog ir lac pud

The result is as below:

Table 5 Result from running regression

3 Variable

s

According to the above result, we now have: X1

General Regression Model:

Sample Regression Model:

^

Or Y =(−2.68e+09)+(3.47e+07)× X^ 1+(5.58e+07)× X2+250977.1 × X3+0.0545 × X4

3 TESTING MODEL

a Multicollinearity

Why is Multicollinearity a Problem?

If the goal is simply to predict Y from a set of X variables, then multicollinearity is not a problem The predictions will still be accurate, and the overall R-squared (or adjusted R-squared) quantifies how well the model predicts the Y values If the goal is

to understand how the various X variables impact Y, then multicollinearity is a big problem:

 One problem is that the confidence intervals on the regression coefficients will be very wide The confidence intervals may even include zero, which means one can’t even be confident whether an increase in the X value is associated with an increase, or a decrease, in Y Because the confidence intervals are so wide, excluding a subject, can change the coefficients dramatically and may even change their signs

 The second problem is that the individual P values can be misleading (a P value can be high, even though the variable is important)

Beside, there are some several other problems can interfere with analysis of results, including:

 The t-statistic will generally be very small and coefficient confidence intervals will be very wide This means that it is harder to reject the null hypothesis

 The partial regression coefficient may be an imprecise estimate; standard errors may be very large

 Partial regression coefficients may have sign and/or magnitude changes as they pass from sample to sample

 Multicollinearity makes it difficult to gauge the effect of indepe ndent variables on dependent variables

Sources of Multicollinearity

There are four sources of multicollinearity:

 The data collection method employed, for example, sampling over a limited range of the values taken by the regression in the population

 Constraints on the model or in the population being sampled

 Model specifications, for example, adding polynomial terms to a regression model, especially when the range of the X variable is small

 An over determined model This happens when the model has more explanatory variables than the number of observations This could happen

in medical research where there may be a small number of patients about whom information is collected on a large number of variables

It is important to understand the differences among these sources of the multicollinearity, as the recommendations for analysis of the data and interpretation of the resulting model depend to some extent on the cause of the problem The data collection method can lead to multicollinearity problems when the analyst samples only a subspace of the region of the regression Constraints of the model can cause multicollinearity An over defined model has more regression variables than number

of observations These models are sometimes encountered in medical and behavioral research, where there may be only a small number of subjects (sample units) available, and information is collected for a large number of regression on each subject The usual approach to dealing with the multicollinearity in this context is to eliminate some of the regression variables from consideration

Effect of Multicollinearity

To assess multicollinearity, it should be noticed that how well each independent (X) variable is predicted from the other X variables And what is the value Variance

Inflation Factor (VIF) When VIF value is high for any of the X variables, the fit is

affected by multicollinearity

We use VIF function to test multicollinearity of the model If at least one of the coefficience has VIF value is greater than 2, we can come to a conclusion that the

model has multicollinearity The result is as follow:

Table 6: Results from running VIF

Mean VIF=1.29 < 2 → Does not have multicollinearity

b Heteroskedasticity

Heteroskedasticity has serious consequences for the OLS estimator Although the OLS estimator remains unbiased, the estimated SE is wrong Because of this, confidence intervals and hypothesis tests cannot be relied on In addition, the OLS estimator is no longer BLUE If the form of the heteroskedasticity is known, it can be corrected

While studying classic linear regression model, we give out a hypothesis that the variance of each Ui in the condition that the given value of explaining variable X is unchanged, which means: u1=u2=…=u i ;i=1,2, … , n

However, in fact, because of the nature of socio-economics, the method of gathering and processing data or the wrong model, the hypothesis is violated causing heteroskedasticity The result of heteroskedasticity is that the minimum estimated square value is not efficient Therefore, the testing is no longer valuable

{ H0: Homoskedasticity

H1:Unrestricted heteroskedasticity

Therefore, if p-value is smaller than 0.05, we reject H0 and accept H1

We use white function to test the model's error.

The result is as below:

Table 7 Results from running Imtest, white

Prob > chi2 = 0.0000 < 0.05, p-value = 0.0000 < 0.05

Hence, we reject H0 and accept H1, then there is heteroskedasticity

c Testing Multiple Linear Regressions: F Test

{ H0: β0=β1=β2=β3=β4=β5=0

H1:thereis at least 1 coefficient ≠ 0

Using test function, the result is as below:

Table 8 Results from running Test

Hence, we accept H0 and reject the regression

d Testing hypothesis in statistic

 Pog: {H0: β1=0

H1: β1≠ 0

o Confidence Interval : (-6.59e+08; 1.35e+09)

o 0 ∈Confidence Interval → Accept H0

o Population growth does not have statistically significant effect on Foreign direct investment

 Lac: {H0: β2=0

H1: β2≠ 0

o Confidence Interval : (31762.09; 470192.2)

o 0 ∉Confidence Interval → Reject H0,accept H1

o Labour cost in one year has statistically significant effect on FDI

 Pud: {H0: β3=0

H1: β3≠ 0

o Confidence Interval : ( -0.0615 ; -0.0475)

o 0 ∉Confidence Interval → Reject H0,accept H1

o Public debt has statistically significant effect on Foreign direct investment

 Ir: {H0: β4=0

H1: β4≠ 0

o Confidence Interval : ( -6.94e+07; 1.86e+08)

o Confidence Interval → Accept H0

o Inflation doesn’t have statistically significant effect on Foreign direct investment.