- CFA Exams
- CFA Level I Exam
- Topic 1. Quantitative Methods
- Learning Module 10. Simple Linear Regression
- Subject 3. Analysis of Variance
CFA Practice Question
Which of the following are true about the SEE (standard error of estimate)?
II. Based on squared vertical deviations between Y and Y'
III. Cannot be negative
I. Measure of the accuracy of the prediction
II. Based on squared vertical deviations between Y and Y'
III. Cannot be negative
Correct Answer: I, II and III
All the descriptions are correct for the standard error of estimate. It is an absolute measure of the fit of the model calculated as the square root of the mean square error.
User Contributed Comments 5
| User | Comment |
|---|---|
| mazen1967 | y bar |
| davidt876 | @mazen: y hat you mean? Also can't the square root of a number be both negative and positive re: answer III? |
| brave1986 | its not the square root but the correlation coefficient raised to the power 2 |
| davidt87 | dude what? this is about SEE not the coefficient of determination. SEE = {E[(Yi - a - bXi)^2/(n-2)]}^(1/2) sq. roots can be negative or positive no? |
| CFAJ | the SEE is a measure of variability away from the estimated values so it would be always be a positive value cos you can't be like negative one foot away from a point if you know what i mean? you are just 1 foot away from that place |