|
Anyone know how to do regression?
|
View this Thread in Original format
| A.J. |
Ok, i know this is desperate, but i was wondering if anyone could help to explain REGRESSION to me.
I have my Business & Economic Statistics Final Exam tomorrow and i have no idea about regression.
I know every other topic perfectly, but regression is one that i haven't really learnt :p |
|
|
| tu_face |
regression is the opposite of progression.
i don't know what exactly it means in an economical sense, but i would take an educated guess at it being something like doing things like you used to, or moving back to old ways or something as such.
| quote: |
1. Reversion; retrogression.
2. Relapse to a less perfect or developed state.
3. Psychology. Reversion to an earlier or less mature pattern of feeling or behavior.
4. Medicine. A subsidence of the symptoms or process of a disease.
5. Biology. The return of a population to an earlier or less complex physical type in successive generations.
6. Statistics. The relationship between the mean value of a random variable and the corresponding values of one or more independent variables.
7. Astronomy. Retrograde motion of a celestial body.
8. Geology. A relative fall in sea level resulting in deposition of terrestrial strata over marine strata.
|
|
|
|
| taltul |
| quote: | Originally posted by A.J.
Ok, i know this is desperate, but i was wondering if anyone could help to explain REGRESSION to me.
I have my Business & Economic Statistics Final Exam tomorrow and i have no idea about regression.
I know every other topic perfectly, but regression is one that i haven't really learnt :p |
there r dif types of regression-methods but in general in statistics regression means determinin' the best relationship between 2 (or more) variables, when u always have 1 (called interpreted or dependent variable) that u r interested in "explainin" it by the other variables (called generally independent variables)... here's simple model: lets say u have data of n observations when each one has 4 variables: e=education, a=age, s=sex and i=income. the possible question about the relation between these variables can be "how much do the e, a and s depend on i?" <=> "whats the relation and the influence of the sex, age and human's education on their income? the formula that u'll get usin regression-method (lets say linear regression in this example) will be something like this:
i = K + L*a + M*e + N*s [linear regression!!!]
whem [K=secant] and [L,M,N=coefficients when each of them explains what is the relation between a e and s to i]. besides there r some other things like mean value or the varience value of regression or the variences of each of the coefficients that u'll need to check their significance (usually with the level of significance equal to 5%)...
the main point is to inderstand if the relation of each independent variable to the dependent 1 is significant enough to rely on the regression's results...
anyway there is a couple of links explainin the linear regression:
1
2
hope u'll find them usefull :)...
enjoy and ask everything u need - i will try my best to helf u...
peace... :) |
|
|
|
|
