Regression And Correlation

Question 1

What is the formula for correlation coefficient?

Answer 1

r = Σxy ÷ (Σx² • Σy²)^½

Where x = X - X̄ ; y = Y - Ȳ
And X̄ = ΣX/N ; Ȳ = ΣY/N

Where N = number of categories (can say columns) NOT sample size

Question 2

When r = +1, what can be inferred?

Answer 2

It is a perfect positive correlation between variables X and Y.

Question 3

When r = -1, what can be inferred?

Answer 3

Perfect negative correlation.

Question 4

When r lies between (or equals) +0.6 and +0.9, what can be inferred?

Answer 4

Highly positive correlation between X and Y.

Question 5

When r lies between (or equals) -0.6 and -0.9, what can be inferred?

Answer 5

Highly negative correlation between X and Y.

Question 6

What is the general equation for method of least squares?

Answer 6

Y = a + bX

Question 7

What are the normal equations which can be used to find least square estimates?

Answer 7

ΣY = na + bΣX

ΣXY = aΣX + bΣX²

Where n = number of categories (can say columns) NOT sample size

Where a and b are constants from the general equation

Y = a + bX.