Difference between Correlation and Regression Analysis

1.Degree and Nature  of Relationship: Coefficient of correlation measures   the degree  of covariance  between two variables  whereas  the regression  analysis  tells  about  the nature  of relationship between the variable  so that one may be  able to estimate or predict the value of one variable  on the basis of another.

2. Cause and Effect Relationship: Correlation mearly  acertains the degree of relationship between two variables  and therefore one cannot  say that one variable  is the cause and other is the effect. In regression analysis one variable is taken as dependent variable while the other variable is taken as independent variable .Thus making it possible to study the cause and effect relationship.

3. The value of rx_y    in the calculation of coefficient of correlation measured the direction and degree of relationship between two variable say x and y. The values of   Ryx   Symmetric , it shows that it is immaterial which  of X   and Y  is dependent variable and which is independent variable. However in the case of regression analysis coefficients i.e.  by xy   are not symmetric i.e.   b _xy * b_yx   and therefore it certainly  makes a difference as to which variable  is dependent and which  one is independent.

4. In case of correlation, there may be nonsense correlation between two variables X and Y which  is due to merely chance and may not have any practical relevance, such as increase in income  and increase of environmental temperature. However there cannot be a nonsense regression. 

5. The value of coefficient of correlation is independent of changes of scale and point of origin. However regression coefficients are independent of changes of origin but not of scale.

6. While pointing out the difference between regression and correlation Werner z .Hirsch rightly stated that. While correlation analysis tests the closeness with which two phenomena co vary, regression analysis measured the nature and extent of the relation, that enabling us to make prediction.

Posted Date: 9/27/2012 6:24:40 AM | Location : United States

Ask an Expert

Related Discussions:- Difference between correlation and regression analysis, Assignment Help, Ask Question on Difference between correlation and regression analysis, Get Answer, Expert's Help, Difference between correlation and regression analysis Discussions

Write discussion on Difference between correlation and regression analysis
Your posts are moderated

Write your message here..

Related Questions

What are the coefficients of the linear combination, For the following ques... For the following questions we are interested in a comparison of the 16 years education vs. > 16 years. (Recall we did the analysis on the log scale, so these are actual means on t

Regression lines, Regression Lines It has already been discussed that t... Regression Lines It has already been discussed that there are two regression lines and they show mutual relationship between two variable . The regression line Yon X gives   th

Arithmetic average or mean, Arithmetic Average or Mean The arithmetic m... Arithmetic Average or Mean The arithmetic mean is the most widely and the most generally understandable of all the averages. This is clear from the reason that when the term

Plot diagnostic quantities, The data in the data frame compensation are fro... The data in the data frame compensation are from Myers (1990), Classical andModern Regression with Applications (Second Edition)," Duxbury. The response y here is executive compens

Define the partial market equilibrium model, Q. The following system of equ... Q. The following system of equations illustrates the algebraic form of a partial (individual) market equilibrium model, which is a model of price (P) and quantity (Q) determination

B) Distinguish between:, X 110 120 130 120 140 135 155 160 165 155 ... X 110 120 130 120 140 135 155 160 165 155 Y 12 18 20 15 25 30 35 20 25 10

Large sample test for mean, Large Sample Test for Mean A random sample ... Large Sample Test for Mean A random sample of size n (n > 30) has a sample mean    . To test the hypothesis that the population mean μ has a specified value  μ 0  let us formu

Advantages of sampling, Advantages of Sampling Why should we settle on ... Advantages of Sampling Why should we settle on a sample instead of studying the entire population?  Sampling has the following advantages over a census (study of the entire pop

Measures of dispersion, calculate variance and standard deviation of the f... calculate variance and standard deviation of the following sample 12,22,32,13,12,23,34,52,56,23,44,32,11,11

Business reporting and analysis, You are a business analyst working for a c... You are a business analyst working for a company called Combined Computers Pty Ltd. You have been asked to prepare a business report with statistics in it for the managing director