This chapter focuses on the finding and data analysis. All the data collected in this study were processed using the SPSS program. SPSS program was used to analyze the data from the correlation and regression analysis. The method was used to analyze the data was Multiple Regression Correlation Analysis. A multiple regression analysis involves more than one independent variable.
The process of evaluating is the same with simple regression, but in order to derive the estimated regression, a computer is employed due to the complex nature of data and time required. The presentation of findings is made to examine the relationship among independent variables (inflation rate, gross domestic product of Malaysia (GDP), interest rate or base lending rate (BLR) and exchange rate) and dependent variable (performance of ASB). Besides that, this research also wants to examine the relationship between profitability (independent variable) and dividend of ASB (dependent variable).
This study used Multiple Regression Method Analysis which is the interpretation of Regression Analysis which includes Coefficient of Correlation (R), Coefficient of determination (R-Square), F-Statistic and T-Statistic.
4.1 REGRESSION EQUATION
General Function:
P = a (INF, GDP, BLR, EXC)
Multiple Regression Equation:
P = - 1.738 - 11703.166 INF -3.912 GDP + 3.428 BLR - 4.813 EXC + e
Where,
P = Profitability
a = constant
b1, b2, b3, b4 = coefficient
INF = Inflation rate
GDP = Gross Domestic Product (GDP)
BLR = Interest rate (BLR)
EXC = Exchange rate
e = Error term
4.2 REGRESSION COEFFICIENT ANALYSIS
Coefficientsa
Model
Unstandardized Coefficients
Standardized Coefficients
t
B
Std. Error
Beta
(Constant)
-1.738E7
3.505E7
-.496
GDP
-11703.166
4655.202
-1.421
-2.514
BLR
-3.912E6
1.764E6
-.324
-2.217
INF
3.428E6
1.121E6
2.076
3.059
EXC
-4.813E6
7.839E6
-.116
-.614
Dependent Variable: Profit
Table 4.1 ( The Regression Result)
4.2.1 RELATIONSHIP BETWEEN DEPENDENT VARIABLE AND INDEPENDENT
VARIABLES
The above table shows the result of coefficient that will be explaining the relationship between dependent variable and independent variables. The positive sign will be explained when the independent variables increasing, dependent variable also will be increasing but for the negative sign will be meant for the opposite relationship, when independent variables increasing, the dependent variable will be decreasing.
4.3 REGRESSION COEFFICIENT INTERPRETATION
4.3.1 CONSTANT
The constant is equal to 1.738 represents the percentage of performance of ASB when the all independent variables which are inflation, gross domestic product (GDP), interest rate (BLR), and exchange rate equal to zero.
4.3.2 GROSS DOMESTIC PRODUCT (GDP)
VARIABLE
REGRESSION COEEFICIENT
GROSS DOMESTIC PRODUCT (GDP)
-11703.166
Table 4.2.1 Table Regression Coefficient between GDP and Performance of ASB
An estimated coefficient of gross domestic product (GDP) is equal to -11703.166. It means, when the gross domestic product increase by 1%, the performance of ASB will be decreasing by 11703.166%. It shows that GDP has negative relationship with the performance of ASB.
4.3.3 INTEREST RATE (BASE LENDING RATE)
VARIABLE
REGRESSION COEEFICIENT
INTEREST RATE (BASE LENDING RATE)
-3.912
Table 4.2.2 Table Regression Coefficient between BLR and Performance of ASB
An estimated coefficient of interest rate (BLR) is equal to -3.912. It means, when the interest rate (BLR) increase by 1%, the performance of ASB will be decreasing by 3.912 %. It shows that interest rate has negative relationship with the performance of ASB.
4.3.4 INFLATION RATE
VARIABLE
REGRESSION COEEFICIENT
INFLATION RATE
3.428
Table 4.2.3 Table Regression Coefficient between Inflation Rate and Performance of ASB
An estimated coefficient of inflation rate is equal to 3.428. It means, when the inflation rate increase by 1%, the performance of ASB will be increasing by 3.428%. It shows that inflation rate has positive relationship with the performance of ASB.
4.3.5 EXCHANGE RATE
VARIABLE
REGRESSION COEEFICIENT
EXCHANGE RATE
-4.813
Table 4.2.4 Table Regression Coefficient between Exchange Rate and Performance of ASB
An estimated coefficient of exchange rate is equal to -4.813. It means, when the exchange rate increase by 1%, the performance of ASB will be decreasing by 4.813%. It shows that exchange rate has negative relationship with the performance of ASB.
4.4 T-STATISTIC (T-TEST)
T-Test is used to determine if there is a significant relationship between the dependent variable and each of the independent variables.
VARIABLES
COMPUTED
T-VALUE
SIGN
CRITICAL
T-VALUE
RESULTS
HYPHOTESIS
PROFIT
-.496
-
-
-
-
GROSS DOMESTIC PRODUCT (GDP)
-2.514
>
2.0
SIGNIFICANT
REJECT HO
INTEREST RATE (BLR)
-2.217
>
2.0
SIGNIFICANT
REJECT HO
INFLATION RATE
3.059
>
2.0
SIGNIFICANT
REJECT HO
EXCHANGE RATE
-.614
<
2.0
INSIGNIFICANT
ACCEPT HO
Table 4.3 Table of T-Statistic Result
Rule of Thumbs T-Statistic = 2
INTERPRETATIONS
4.4.1 GROSS DOMESTIC PRODUCT (GDP)
Since Rule of Thumb is 2, therefore at 95% confidence level, the calculated T value is more than critical T value from the distribution table (2.514 > 2.000). Therefore there is significant relationship between the profitability and gross domestic product. So, from the hypothesis, H0 will be rejecting and H1 will be accepted. This finding is supported by Robert E. Hall and Marc Lieberman (2001), when the economy is expanding, real GDP is rising, firms is general tend to earn high profits, and these profits are less risky. This is in addition to the normal rise in real GDP that would be occurring anyway, as income growth. In the typical expansion, profits will rise along with GDP. Higher profits are enough to make stocks look more attractive.
4.4.2 INTEREST RATE (BLR)
Since Rule of Thumb is 2, therefore at 95% confidence level, the calculated T value is more than critical T value from the distribution table (2.217> 2.000). Therefore there is significant relationship between the profitability and interest rate. So, from the hypothesis, H0 will be rejecting and H1 will be accepted. This finding is supported by Lawrence J. Gitman Jeff Madura (2001), the changes in interest rate affect consumer's purchases with borrowed funds and the firm's cost of financing. Most of the firms are unfavorably affected by upward movements in interest rates and are favorably affected by downward movements in interest rates.According to Gonzales et al. (2000), the interest rate are a proxy for the stance of monetary
policy and this is why it is reasonable to think that they could predict the stock returns. The spread between long and short term interest rates is also the stance for the monetary policy. Long term yields contain a risk premium above the average of expected future short term yields. A sign of monetary ease is when long term yields are high than short term yields.
4.4.3 INFLATION RATE
Since Rule of Thumb is 2, therefore at 95% confidence level, the calculated T value is more than critical T value from the distribution table (3.059> 2.000). Therefore there is significant relationship between the profitability and inflation rate. So, from the hypothesis, H0 will be rejecting and H1 will be accepted. This finding is not supported by Chen et al. (1986). They used monthly data for the period 1958 to 1984 to test the impact of the inflation rate on stock prices. In fact, they defined three variable related to the inflation rate: expected inflation; the change in expected inflation; and unanticipated inflation, and found a significantly negative relationship between the inflation and stock prices. Besides that, Geske and Roll (1983) and Chen, Roll and Ross (1986) in their research show a negative relationship between inflation and equity returns. This is same with later result from Murkherjee and Naka (1995) that show a negative relationship between Tokyo Stock Exchange and inflation.
4.4.4 EXCHANGE RATE
Since Rule of Thumb is 2, therefore at 95% confidence level, the calculated T value is less than critical T value from the distribution table (.614<2.000). Therefore there is insignificant relationship between the profitability and exchange rate. So, from the hypothesis, H1 will be rejecting and Ho will be accepted. According to Aggarwal (1981) stated that the relationship between exchange rates and stock prices using monthly data from 1974 to 1978 by using correlation regression analyses. The study found that the trade-weighted exchange rate and the stock market indices were positively correlated during this research period. Movement of exchange rate could directly affect the stock prices of multinational firms by influencing the value of its overseas operations, and indirectly effect domestic firms through influencing the prices of its exports or imported inputs. Moreover, according to Ma and Kao (1990), they stated that relationship between exchange rates and the stock prices in six industrialized economies, the U.K, Canada, France, West Germany, Italy and Japan using monthly data from January 1973 to December 1983. They tested the degree of stock price reaction to exchange rate changes and their findings were consistent with the exchange rate movement caused the stock price movement thus will affect the unit trust as whole.
4.5 F-STATISTIC (F-TEST)
An F-test is a statistical test which most used when comparing statistical models that have been fit to a data set, in order to identify the model that best fits the population from which the data were sampled, Exact F-Test always arise when the models have been fit to the data using least squares.
Computed F-Stat
Critical F-Value
Interpretation
23.383
4 (RULE OF THUMBS)
Significant
Table 4.4 Table of F-Statistic Result
F-Statistic = 4 (RULE OF THUMBS)
Lastly, by looking at the F -Stats, the value is equal to 23.383. This shows that the calculated F value is higher than the value of F in table (23.383>4). It means that the data taken is reliable for the overall model. This data also provide strong evidence to evaluate the significant of each component of the entire model. So, from hypothesis, H1 will be accepted and H0 will be rejected.
4.6 COEFFICIENT OF DETERMINATION (R2)
R2 is the coefficient of determination where it is used to test the goodness of fit. It measure how many % of a change / variation in the dependent variable and be measured or explained by in the independent variables. The coefficient of determination, R2, is useful because it gives the proportion of the variance (fluctuation) of one variable that is predictable from the other variable. It allows us to determine how certain one can be in making predictions from a certain model/graph.
Model Summary
Model
R
R Square
Adjusted R Square
1
.928a
.862
.825
Predictors: (Constant), Exc, Interest, Gdp, Inflation
Table 4.5 Table of R-Squared
From the above result, it shows that R-squared (R²) is 0.862. It can be considered as giving a very high explanatory power for the estimated equation. In other words, it means 86.2% of the change in the variables can be explained by independent variables but only 13.8% can be explained by others factor. Adjusted R-squared is a modification of R² that adjusts for the number of explanatory terms in a model.
4.7 LINEAR REGRESSION EQUATION
(PROFITABILITY AND DIVIDEND OF ASB)
DIV = -10.00 - 2.79 P + e
Where,
P = Profitability
a =constant
b1 =coefficient
DIV = Dividend of ASB
e = error term
4.8 REGRESSION COEFFICIENT ANALYSIS
(PROFITABILITY AND DIVIDEND OF ASB)
Coefficientsa
Model
Unstandardized Coefficients
Standardized Coefficients
t
B
Std. Error
Beta
1
(Constant)
10.001
.671
14.894
Profit
-2.794E-7
.000
-.556
-2.838
a. Dependent Variable: Dividend
Table 4.6 Table of Regression Result
Relationship between dependent variable and independent variables.
The above table shows the result of coefficient that will be explaining the relationship between dependent variable and independent variables. The positive sign will be explained when the independent variables increasing, dependent variable also will be increasing but for the negative
sign will be meant for the opposite relationship, when independent variables increasing, the dependent variable will be decrease
4.9 REGRESSION COEFFICIENT INTERPRETATION
4.9.1 CONSTANT
The constant is equal to 10.001 represents the percentage of dividend of ASB when the independent variables which are profitability equal to zero.
4.9.2 PROFITABILITY
VARIABLE
REGRESSION COEEFICIENT
PROFITABILITY
-2.794
Table 4.7 Table Regression Coefficient between Profitability and Dividend of ASB
An estimated coefficient of profitability is equal to -2.794. It means, when the profitability increase by 1%, the dividend of ASB will be decreasing by -2.794%. It shows that profitability has negative relationship with the dividend of ASB.
5.0 T-STATISTIC (T-TEST)
T-Test is used to determine if there is a significant relationship between the dependent variable and each of the independent variables.
VARIABLES
COMPUTED
T-VALUE
SIGN
CRITICAL
T-VALUE
RESULTS
HYPHOTESIS
DIVIDEND OF ASB
14.894
-
-
-
-
PROFITABILITY
-2.838
>
2.0
SIGNIFICANT
REJECT HO
Table 4.8 Table of T-Statistic Result
RULE OF THUMBS T-STATISTIC = 2
INTERPRETATION
4.9.1 PROFITABILITY
Since Rule of Thumb is 2, therefore at 95% confidence level, the calculated T value is more than critical T value from the distribution table (2.838> 2.000). Therefore there is significant relationship between the dividend of ASB and profitability. So, from the hypothesis, H0 will be rejecting and H1 will be accepted.
5.1 F-STATISTIC (F-TEST)
An F-test is a statistical test which most used when comparing statistical models that have been fit to a data set, in order to identify the model that best fits the population from which the data were sampled, Exact F-Test always arise when the models have been fit to the data using least squares.
Computed F-Stat
Critical F-Value
Interpretation
8.054
4 (Rule Of Thumbs)
Significant
4.9 Table of F-Statistic Result
F-Statistic = 4 (Rule of Thumbs)
Lastly, by looking at the F -Stats, the value is equal to 8.054. This shows that the calculated F value is higher than the value of F in table (8.054>4). Therefore, there is significant relationship between the independent variables and dependent variable. It means that the data taken is reliable for the overall model. This data also provide strong evidence to evaluate the significant of each component of the entire model. So, from hypothesis, H1 will be accepted and H0 will be rejected.
5.2 COEFFICIENT OF DETERMINATION (R2)
R2 is the coefficient of determination where it is used to test the goodness of fit. It measures how much percentage of a change / variation in the dependent variable and be measured or explained by in the independent variables. The coefficient of determination, R2, is useful because it gives the proportion of the variance (fluctuation) of one variable that is predictable from the other variable. It allows us to determine how certain one can be in making predictions from a certain model/graph.
Model Summary
Model
R
R Square
Adjusted R Square
1
.556a
.309
.271
a. Predictors: (Constant), Profit
5.0 Table of R-Squared
From the above result, it shows that R-squared (R²) is .309. It is very low and can be considered as not very high explanatory power for the estimated equation. In other words, it means 30.9% of the change in the variables can be explained by independent variables and 69.1% can be explained by others factor. Adjusted R-squared is a modification of R² that adjusts for the number of explanatory terms in a model.
