Correlation Analysis Using Ordinary Least Square Test

CHAPTER 4

DATA ANALYSIS

Coefficient Correlation Analysis

The first analysis is by using Ordinary Least Square (OLS) test to measure the relationship of entire variables. The test is to find the function which most closely approximates the data. Thus, in general terms, it is an approach to fitting a model to the observed data. The details information regarding the variables is shown in table 4.1 and table 4.2 shows the least square test that measures all the variables.

Variables Description

LGP Log Gold Price (MYR/oz)

LCPO Log Crude Oil Price (MYR/barrel)

LEX Log Exchange Rate (MYR/USD)

LIR Log Interest Rate (%)

Table 4.1

Dependent Variable: LGP

Method: Least Squares

Date: 06/22/15 Time: 12:52

Sample: 1 84

Included observations:

Error t-Statistic Prob.

C 4.450579 0.571797 23.52333 0.0000

LCPO 0.263956 0.071989 3.666637 0.0004

LEX 0.730580 0.253322 11.17382 0.0000

LIR -0.4192 0.087159 -2.916415 0.0046

Table 4.2

R-squared 0.609579 Mean dependent var 8.356235

Adjusted R-squared 0.594939 S.D. dependent var 0.169791

S.E. of regression 0.108063 Akaike info criterion -1.565763

Sum squared resid 0.934204 Schwarz criterion -1.450009

Log likelihood 69.76203 Hannan-Quinn criter. -1.519231

F-statistic 41.63574 Durbin-Watson stat 0.597958

Prob(F-statistic) 0.000000

Correlation coefficient test is a test used to identify the strength and direction of the linear relationship between variables. The objective is measure the correlation between the dependent variable, gold price in Malaysia with at least 1 of these independent variables which consist of crude oil price, exchange rate and interest rate in Malaysia. This correlation needs to be done primarily in order to perform the Multiple Regression Model analysis.

Based on E-Views method, there are 3 values where each of it stipulates the correlation coefficient, R-squared, probability of F-statistics and p-value of t-test

Error t-Statistic Prob.

C 4.450579 0.571797 23.52333 0.0000

LCPO 0.263956 0.071989 3.666637 0.0004

LEX 0.730580 0.253322 11.17382 0.0000

LIR -0.254192 0.087159 -2.916415 0.0046

Table 4.2.1

4.3 REGRESSION OUTPUT MODEL

This equation for the hypotheses of this research could be analyzed and also to figure out how the relationship can affect each other between dependent and independent variables.

The equation can be derived as follows:

Where: Y= 4.450579+ 0.263956β1 + 0.730580β2 -0.254192β3 + Σ (0.571797) (0.071989) (0.253322) (0.087159)

From the equation above, it shows that only crude oil price has significant positive relationship with gold price in Malaysia. Adversely, the remaining variable exchange rate and interest rates in Malaysia have negative relationship with gold prices. Crude oil = 0.263956

From the equation, we can interpret that crude oil price have a positive relationship with gold prices. It indicates that the two variables are positively correlated while holding other variables constant. Crude oil will increase by 0.2639%, this also means that for any one unit increase or changes in crude oil price, gold prices will also increase by