correlation and regression

Question 1

How to choose between regression and correlation?

Answer 1

• c = interdependent variables
• regression = normally distributed data - dependent and independent variables

Question 2

What is the difference betwen correlation and regression?

Answer 2

C = quantifies degree to which two variables are related
R = linear r finds best line that predicts Y from X
C = Correlation coefficient = how much one variable tends to change when the other one does
R= Casual
C = Doesn’t matter which is X and which is Y
C= calculates a correlation coefficient (R) from -1 to +1
R = calculates goodness of fit (r^2)

Question 3

What are the choices for correlation?

Answer 3

• Parametric (normal distribution = pearson) - one variable increases as the other increases
• non parametric = spearman
• one variable decreases as the other increases

Question 4

what is the pearson correlation coefficient r?

Answer 4

• determines whether there is a statistically significant correlation
• the sign of r (+ or -) indicates whether there is positive or negative correlation
• the closer r is to 1 , the stronger the correlation

Question 5

Regression-linear

Answer 5

• in regression analysis, the distance between the data point and the corresponding point on the line of best fit is minimised for all of the data points
• it assumes normal distribution for the values and errorrs of the y variable
• there is no error in the x variable
• the variability of the errors for the y varibale are constant for all x variables