Chapter 3

METHODOLOGY

3.1

Introduction

This chapter

consists with seven parts such as Research Framework or Model of the Analysis,

Hypotheses Development,

Research Design,

Operational Definitions,

Measurement of Variables,

Data Collection & Sampling,

Data Collection Procedures,

and Techniques of Data Analyses.

3.2 Research Framework:

Y= GDP per capita

X2=G.

N. Savings

X1=

Inflation

X4=

Remittance

X3=

Export

3.3 Model:

Y = ? + ?1X1 +

?2X2 + ?3X3 +

?4X4 + € ,

Where,X1 shows the inflation rate , X2 shows the Gross

national savings, X3 shows the Export and X4 Shows Remittance of Bangladesh . All

data of this model are collected annual basis.

3.4

Hypotheses Development

Y

> 0; self-dependent GDP per capita

when inflation, Gross national savings,

Export and remittance are zero though it is not very

important.

?1>0If

inflation increases, then

the GDP per capita will be increased.

?2>0If

savings increase, then

the GDP per capita will be increased.

?3>0

If

export increases,

then the GDP per capita will be increased.

?4>0

If

remittance

increases, then the GDP per capita will be increased.

3.5

Measurement of Variables

In order to examine the impact of

inflation, Gross national savings, Export and remittance on GDP per capita, we

have specified following econometric model. The independent variables are inflation,

Gross national savings, export and remittance, while the dependent variable is

economic GDP per capita. The model is stated as follows :

Y = ? + ?1X1 +

?2X2 + ?3X3 +

?4X4 + € ,

GDP per capita income= f(Inf.r,

Sav.r, Exp.r, Rem.r,)

In linear form,

equations can be written as:

GDP per capita income

=f(?1 Inf.r+ ?2 Sav.r+ ?3Exp.r + ?4Rem.r)

3.6

Operational Definitions

In this paper I

am collecting annual based time series data, from various source and tabulate data

in ms excel year by year and try to find out the relationship among the various

annual based data. Then set econometric model , among this data we set GDP per

capita income as a dependent variable(which is shown by US$), and all others

are independent variable which are influence the GDP per capita income of

Bangladesh .after getting result we see all variable are very important for GDP

per capita income .

3.7

Data Collection, Procedures and Sampling

Data were

collected from the secondary sources and taken the period of 1976-2016, this

means that we have 41 years data (n=41). And this data collected from based on

web-side of WB, BB, IMF and from

various economic review yearly book. For robustness of analysis, variables under

consideration are economic GDP per capita income (proxied by gross domestic

product), inflation rate, gross national

savings rate, Export and remittance of Bangladesh.

3.8 Techniques of Data Analyses

All data have

been tabulated by sequence number for final use. MS Excel 2007 and SPSS

software version 12, use also stata 10.0are used to reach the destination in

this project. To make a proper regression we need to cheek the Descriptive

statistics, correlation analysis ,ANOVA, model summery, Regression Analysis for getting proper result.

Chapter 4

FINDINGS

OF THE STUDY

4.1 Introduction

This chapter is very

important of my study paper. In this chapter first explain descriptive summery

, then find-out the correlation among the variables , 5 percent significant

level among the variables and overall model discussion.

Table 4.1.1 :Descriptive Statistics

N

Minimum

Maximum

Mean

Std. Deviation

Year

41

1976.00

2016.00

1.9960E3

11.97915

Y

41

128.80

1358.70

4.4597E2

297.36092

X1

41

-17.60

25.60

6.8195

6.50601

X2

41

-2.90

24.90

13.8585

6.94863

X3

41

.50

36.80

8.7356

10.49943

X4

41

.02

15.20

3.9512

4.96175

Valid N

(listwise)

41

In the above table, there are five group

of data set( GDP per capita US$ , Inflation rate , Gross national savings as

percent of total GDP of year , export and remittance are collected by billions of dollars). The mean

value of GDP per capita, inflation rate, national savings , export earnings and

remittance is 445.69, 6.81, 13.85, 8.73 and 3.95 respectively .

Table 4.1.

2:Correlations

GDP(US$)

Inflation.R

G.savings

Export

Remittance

GDP(US$)

1

41

Inflation.R

-.024

1

41

G.savings

.765**

-.110

1

41

Export

.988**

-.042

.733**

1

41

Remittance

.957**

-.028

.714**

.981**

1

41

Note: **Significant at .01

level

In the above table shows the correlation among five

variables. (GDP per capita, inflation rate , gross national savings , export

and remittance). The total number of observations are 41, the correlation between GDP per

capita and inflation rate is -0.0240 correlation coefficient , the correlation

between GDP per capita and gross savings is

0.7654 , and 0.9880 and 0.9567

correlation between GDP per capita and export, and correlation between

GDP per capita and remittance respectively .

Table: 4.1.3: ANOVAb

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

3483854.888

4

870963.722

590.641

.000a

Residual

53085.881

36

1474.608

Total

3536940.769

40

a.

Predictors: (Constant), X4, X1, X2, X3

b.

Dependent Variable: Y

The one-way ANOVA compares the means between the groups you are

interested in and determines whether any of those means are statistically

significantly different from each other. Specifically, it tests the null

hypothesis:

H0:µ1 = µ2 = µ3= …= µK

where µ = group mean and k =

number of groups. If, however, the one-way ANOVA returns a statistically

significant result, we accept the alternative hypothesis (HA), which

is that there are at least two group means that are statistically significantly

different from each other.

The SPSS output for the ANOVA is shown in the above

table, indicating whether we have a statistically significant difference

between our four group means. We can see that the significance level is 0.00 (p =

.00), which is below 0.05. and, therefore, there is a statistically significant

difference in the mean productivity between the four different groups of the independent

variable, (inflation rate ,

Gross National savings, Export , Remittance).

Table 4.1.4 : Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.992a

.985

.983

38.40062

a.

Predictors: (Constant), X4, X1, X2, X3

From t-statistics and probability value we

can test (inflation rate

, Gross National savings, Export,

Remittance). Significantly affection GDP per capita (US$) or not.

Now, we can set null and alternate hypothesis

as follows-

H0: ? 1=0 : ? 1 ?0

and

H0: ? 2=0 : ? 2?0

Here,

we see that the probability value of coefficient of (inflation rate , Gross National

savings, Export , Remittance)are

.168, 0.005,0.00 and .002 respectively,

which without inflation rate all are less than 0.05 where the level of significance

is 5%. We can reject null. Only inflation rate is more than .05% so only for X1

we can accept null hypothesis.

In our study we consider the following model

because of economic significance. This study proceeds with the OLS method.

Y = ? + ?1X1

+ ?2X2 + ?3X3 +

?4X4 + €

Estimated results with Ordinary Least Square

method has been reported in

Table : 4.1.5 : Coefficientsa

Model

Unstandardized Coefficients

Standardized Coefficients

t

Sig.

B

Std. Error

Beta

1

(Constant)

152.612

16.727

9.124

.000

X1

1.325

.942

.029

1.406

.168

X2

3.895

1.293

.091

3.012

.005

X3

35.826

3.103

1.265

11.545

.000

X4

-20.910

6.391

-.349

-3.272

.002

a.

Dependent Variable: Y (GDP US$)

In

the above table hare we calculate the statistically significance about

Inflation rate, gross national savings , remittance and export on GDP per

capita income (US$) of Bangladesh .As the p-value is much more

than 0.05, we accept the null hypothesis that ? = 0. In the above model there are four variables. First one

is X1 ( inflation rate ) which is statistically insignificant because the

result is more than 0.05 percent .here shows the value is .689, so the

relationship between inflation rate and GDP in this model is not

significant.

Second variable is X2(Gross national savings) which is

statistically significant because the result is .005 percent because the result

is less than .05. So the relationship between Gross national savings and GDP in

this model is significant. Third

variable is X3(Export) which is also statistically significant because the

result is .00 percent which is fully significant, so the relationship between export

rate and GDP in this model is positive and also significant. Second variable is X4(remittance) which is

statistically highly significant because the result is .002percent. So the

relationship between remittance and GDP in this model is positive and also significant. We can see

that the significance level is 0.015 (p = .015), which is above

0.05. and, therefore, there is a statistically insignificant difference in the

mean productivity between the four different groups of the independent variable.

In this model we can see.

4.2R2 vs.

Adjusted R2

R2 Measures the proportion of the variation in the GDP

that is explained by variations in the (inflation rate, Gross National savings,

Export , Remittance). In our regression model, 0.985% of the total variation

was explained. Adjusted R2 is a measure the proportion

of the variance in the GDP that is explained by variations in the (inflation

rate , Gross National savings, Export, Remittance). Here, our regression model

shows that 0.983% of the variance is explained. It is need to report because it

‘corrects’ for adding more variables to a regression. Adjusted R2 indicates

that if we add more explanatory variables, it will lead greater R2.

So, it is reasonable that adjusted R2