Page 175 - Contributed Paper Session (CPS) - Volume 8

CPS2228 Mohd Zulhairi Omar et al.
aims to produce an empirical evidence for showing the relationship between
number of employees and salaries in manufacturing sectors from June 2017
to June 2018 without excluding the outlier value.

2. Methodology
In this study, the measure of correlation coefficient was computed on a
real data set by using Pearson’s correlation coefficient. Let (x1,y1), …, (xn,yn) be
n observations from a bivariate normal distribution with parameters (µx,µy,σ x,
2
2 2 2
σ y), where µx and σ x are the mean and variance of x, µy and σ y are the mean
and variance of y.
The Pearson’s correlation coefficient r is the define as:

For robust correlation coefficient measure, this study used the robust Median
Absolute Deviation (MAD) introduced by Gideon (1987, 2007) to calculate
correlation coefficient. This method is very general and provides a robust
estimation procedure in basic correlation analysis and in every advanced
statistical procedure.

This study used the data from Monthly Manufacturing Report that be
published by Department of Statistics Malaysia (DOSM) on June 2018. This
survey covers 155 out of a total of 259 industries in the Manufacturing Sector
(based on the Malaysia Standard Industrial Classification, 2008).

164 | I S I W S C 2 0 1 9