Hi, I’m really having trouble with this subject. This question I can’t figure out, even though I have the anser. A utility analyst performed a regression analysis relating monthly energy consumption to average monthly temperature over the last four years. Total variation of the dependent variable was 58.6, and the unexplained variation was 31.3. What is coefficient of determination and standard error of the estimate?
The answers are: The coefficient of determination is explained by variation divided by total variation. (58.6 - 31.3) / 58.6 = .8249. Schweser questions don’t ask how to compute the coef of determination, so I’m a little confused. Standard error of the estimate: There are 48 observations in the sample: the standard error of estimate is the square root of (31.3 / 46) = .8249 Can anybody tell me how there are 48 observations…? and how the formula is 31.3 / 46…I thought the SEE was the sum of the squared errors the number of observations. thanks
Here Temperature (T) is the Indep var and MEC is the Dep variable Coeff of Determination (R^2) is the amount of variation in the dependent variable explained by the indep variable Therefore Mathematically, R^2 = amount of variation explained / total variation from the problem, it’s given that… TV = 58.6, UEV = 31.3 TV = EV + UEV EV = 58.6 - 31.3 = 27.3 therefore R^2 = 27.3/58.6 = 0.46587 Now, SEE = SQRT(unexplained variation / n-2) = SQRT(31.3/ 46) =0.82488 are these answers correct??? - Dinesh S EDIT: oops sorry, I didn’t read your previous post… It took am time to type this in…
Dinesh, how do we know that there are 48 observations?
Monthy observation for 4 years would fetch us 12*4 = 48 observations - Dinesh S
dinesh, thanks for your help. I have to run to church, then lunch with wife, mother in law, sisters in law…birthday lunch. then I will return to Regression. If you do not mind, I may have more questions later. thanks yancey
bbye,enjoy you lunch and don’t forget to get me a to-go chicken schezwan.
I don’t know about you…but, while having lunch with my mother in law & sister in law, is the only time I would look forward to Regression.