relationships between two or more than two variables is studies is called CORRELATION. Correlation Coefficient Measures the degree to which the change in one variable (the dependent variable) is associated with change in the other variable (independent one). Once we know that two variables are closely related, we can estimate the value of one variable, given the value of other. Correlation and Causation Correlation analysis helps us in determining the degree of relationships between two or more variables – It does not tell us anything about “Cause – Effect” relationship. Correlation does not necessarily imply causation of functional relationship though the existence of causation always implies correlation. Correlation …Contd. Month Advtg. Exp. Sales Rs. The following are the January 50 1200 monthly figures of February 60 1500 advertising expenditure March 70 1600 and sales of a firm. It is April 90 2000 May 120 2200 generally found that ad. June 150 2500 expenditure has its impact July 140 2400 on sales generally after August 160 2600 two months. Allowing for September 170 2800 this time lag, calculate co- October 190 2900 November 200 3100 efficient of correlation. December 250 3900 Correlation …Contd. Solution: Allow for a time lag of 2 months i.e. link ad. Expenditure of January with sales for March and so on. Month Ad. Exp. x2 Sales y2 xy X Y x y January 50 -7 49 1600 -10 100 70 February 60 -6 36 2000 -6 36 36 March 70 -5 25 2200 -4 16 20 April 90 -3 9 2500 -1 1 3 May 120 0 0 2400 -2 4 0 Correlation …Contd.
Month Ad. Exp. x2 Sales y2 xy
X Y x y June 150 +3 9 2600 0 0 0 July 140 +2 4 2800 +2 4 4 August 160 +4 16 2900 +3 9 12 September 170 +5 25 3100 +5 25 25 October 190 +7 49 3900 +13 169 91 ∑1200 ∑x = 0 ∑x2 ∑y ∑y = 0 ∑x2 ∑xy = 261 = 222 = 26,000 Correlation …Contd. = = 120 = = 2600
log r = log 261 – ½ (log 222 + log 364)
= 2.4166 – ½ (2.3464 + 2.5611) = 2.4166 – 2.4537 = T .9629 = ALT .9629 r = 0.918