CORRELATION AND REGRESSION ANALYSIS ON SAVINGS AND LOANS
CHAPTER ONE
1.0 INTRODUCTION
The term regression was originally used by FRANCIS GALTON (1822 – 1911) in a statistical examination of human inheritance to denote certain hereditary relationship very oen in practice, a relationship is found to exist between these variables and oen this make it possible to predict one or more variables in terms of others. For instance, studies are made to predict the future sales of a new product with respect to its price, family
expenditure on feeding in terms of the family income, the consumption of certain food items in relation to the amount spend on its advertisement, quality of a product depending on the temperature of the product at production etc. In this study, the focus is to examine the relationship between savings and loans with reference to EDE COMMUNITY MICRO FINANCE BANK NIGERIA LIMITED (ECB).
The group of data to be used involved only two variables; savings and loans, hence the simple linear regression and correlation analysis shall be used. The saving is taken to be independent variable (X) on which the loans (Y) depend. The statistical model for sample linear regression assumes that for each value of X, the observed value of the response variable Y are normally distributed about a mean that depends on X. we use my to represent these means. Rather than just two m1 and m2, we are interested in how many means my changes as X changes. In general the my changes according to any sort pattern as X changes. In linear regression, we assume that they all line on a line when plotted against X. the equation of the line is given by
Y = a + bx
Where,
y = dependent variable
a = constant parameter which is autonomous (intercept)
b = parameter which shows the rate of change y with respect to x (slope).
With intercept a and slope b; this is the regression line. It describes how the mean response change with x. actually observed y’s will vary the mean. The model assumes this variation measures by the standard deviation r is the same for all the value of x. the response y to a given x is a random variable that will take different
value if we have several observations with the x – value.
The strength of this relationship is determined by the amount of effect that any change in one variable may cause on the other hand.
