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

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
Student’s name

Student ID

Nguyễn Thị Vân

1812250041

Hoàng Phương Linh

1817250021

Nguyễn Thị Như Quỳnh

1816250033

Lê Ngọc Khánh Huyền

1812250019

Nguyễn Quỳnh Giang

1812250015

Lecturer: PhD. Dinh Thi Thanh Binh

Hanoi, 2019

LUAN VAN CHAT LUONG download : add


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

2

LUAN VAN CHAT LUONG download : add


3

LUAN VAN CHAT LUONG download : add


INTRODUCTION
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!

4

LUAN VAN CHAT LUONG download : add


CONTENT
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.

5

LUAN VAN CHAT LUONG download : add


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 :

II.



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



Part 2 : Analyze regression model and correlation



Part 3 : Testing model



Part 4 : Conclusion

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)

6

LUAN VAN CHAT LUONG download : add




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.

7

LUAN VAN CHAT LUONG download : add


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

8

LUAN VAN CHAT LUONG download : add


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 (X1 ), ir (X2 ), lac (X3 ), pud
(X4 ) has the following form:
General Regression Model FDI =β 0 + β 1 × pog+ β 2 ×ir + β 3 ×lac+ β 4 × pud + u^ i
Sample Regression Model
^

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

9

LUAN VAN CHAT LUONG download : add


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

10

LUAN VAN CHAT LUONG download : add


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. Variables

Coefficients

Values

t

p-values

β0

-2.68e+09

-1.7

0.092

Pog

β1

3.47e+07

0.68

0.496

Ir

β2

5.85e+07

0.91

0.367

lac

β3

250977.1

2.27

0.025

pud

β4

-0.0545

-15.48

0.000

According to the above result, we now have: X1
General Regression Model:
FDI=(−2.68e+09)+(3.47e+07) × pog+(5.58e+07)×ir +250977.1× lac -0.0545 × pud + u^ i

Sample Regression Model:
^
FDI=(−2.68e+09)+(3.47e+07) × pog+(5.58e+07)×ir +250977.1× lac +0.0545 × pud + u^ i

11

LUAN VAN CHAT LUONG download : add


Or Y^ =(−2.68e+09)+(3.47e+07)× X 1+(5.58e+07)× X 2 +250977.1 × X 3 +0.0545 × X 4
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 independent
variables on dependent variables.

Sources of Multicollinearity
There are four sources of multicollinearity: 

12

LUAN VAN CHAT LUONG download : add




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.

13

LUAN VAN CHAT LUONG download : add


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

14

LUAN VAN CHAT LUONG download : add


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=u 2=…=ui ; 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.

{

H 0 : Homoskedasticity
H 1 :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:

15

LUAN VAN CHAT LUONG download : add


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

{

H 0 : β 0=β 1=β 2= β3 =β 4=β 5=0
H 1 :there is at least 1 coefficient ≠ 0

Using test function, the result is as below:

Table 8. Results from running Test
16

LUAN VAN CHAT LUONG download : add


p−value=0.1348>0.05

Hence, we accept H0 and reject the regression.
d. Testing hypothesis in statistic

{

H 0 : β 1=0

 Pog: H : β ≠ 0
1
1

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.

{

H 0 : β 2=0

 Lac: H : β ≠ 0
1
2

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.

{

H 0 : β 3=0

 Pud: H : β ≠ 0
1
3

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.

{

H 0 : β 4 =0

 Ir: H : β ≠ 0
1
4
o Confidence Interval : ( -6.94e+07; 1.86e+08)
o Confidence Interval  → Accept H

0

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

17

LUAN VAN CHAT LUONG download : add


IMPLICATION AND CONCLUSION
As a whole, after working on chosen database and analyzing, our group suggest the
best model to quantify the effects of some factors contributing to FDI. 
In addition, to expand further, so how to increase in FDI of a country ? or What
should  the government do to improve FDI ? 
Firstly, based on a review of the latest research on the role of population in
economic growth and the determinants of FDI, it was hypothesized that a country's
population would be positively related to FDI. 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. So, to increase in
FDI of a country, we also need to increase in population .
Secondly, when we decrease inflation rate, the FDI will increase because
inflation rate and foreign direct investment are adversely related. In other words with
an increase in the inflation rate the inflow of foreign investment decreases. The
negative relationship between inflation and foreign direct investment is due to the fact
that high level of prices in the country results in rising production costs. This is due to
the increase in input prices, cost of raw material, wages of labor, land prices and cost
of capital. Such high prices of product also adversely affects domestic as well as
foreign demand for commodities.
Thirdly ,to increase in FDI , we should decrease in public debts .when public
debts is low, it will ensure the economy will develop stably and benefit when
investing, then foreign investment will be effective, so FDI will increase.

Fourthly, the government should open markets and allow for FDI inflows.
Reduce restrictions on FDI. Provide open, transparent and dependable conditions for
all kinds of firms, whether foreign or domestic, including: ease of doing business,
access to imports, relatively flexible labour markets and protection of intellectual
property rights.
Fifthly ,the government should set up an Investment Promotion Agency (IPA).
A successful IPA could target suitable foreign investors and could then become the
18

LUAN VAN CHAT LUONG download : add


link between them and the domestic economy. On the one side, it should act as a onestop shop for the requirements investors demand from the host country.
Finally, put up the infrastructure required for a quality investor: such as
sufficient close-by transport facilities (airport, ports), adequate and reliable supply of
energy, provision of an adequately skilled workforce, facilities for the vocational
training of specialised workers, ideally designed in cooperation with the investor
(Ibid.).
Our report was made based on the contributions of members with knowledge gained
from the process of studying and studying Econometrics. This is also a hands-on
opportunity for us to better understand the relevant analytical and characteristic tests,
apply the knowledge in the lecture hall to self-learn and draw useful conclusions about
phenomena in the various correlations and interplay between socio-economic
phenomena. Our team has completed the econometric model on the factors that affect
FDI data. Thus, through the results of the regression model, our team found that this
data set when being included in the model was not really appropriate, the explanation
of the independent variables was very low, and there were many errors in the process
perform .
Through this, my group also sincerely thanks for the guidance and enthusiastic
teaching of Ms. Dinh Thi Thanh Binh. Due to the limited knowledge and skills, this

report cannot avoid mistakes, we are looking forward to receiving your encouraging
suggestions so that we can improve and apply better in later jobs in the future.

19

LUAN VAN CHAT LUONG download : add