| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Stop wasting time looking for files and revisions. Connect your Gmail, DriveDropbox, and Slack accounts and in less than 2 minutes, Dokkio will automatically organize all your file attachments. Learn more and claim your free account.

View
 

lec0

Page history last edited by PBworks 12 years, 8 months ago

questions from lec0 survey:

 

1.(I don't know if this is the place I should ask a general question from the reading...If not, please let me know.  I can bring this question to the office hour.)  One question from reading.  On page 2, the textbook talks about the use of 'jittering' to avoid 'overplotting' with many points at exactly the same location.  I don't understand how 'jittering' would be the solution.  Maybe I have this question because I don't quite understand the concept of 'jittering'  Could you explain the concept of 'jittering'?

 

T: jittering means adding a little noise (like the small shaking of a hand when you are holding something so multiple points on the same dot get spread out a little

2.I think I just need to review regression from last semester so I can feel more comfortable.

 

T: we are going to start regression in lec1...I don't assume that you did any regression last semester.

 

3. The linear regression equation, wages=b0+b1sex, I did not understand why you plugged the male number in for

b0 and female in for b1.  Why did you plug in 0 for sex?  How do you know which value to assign in which place? 

I'm not sure if I understood exactly what you were asking us to do with the equation.

 

Response (T): the coding of 0 for male and 1 for female is arbitrary.  I.e., I could have used any coding. 

This would affect the values of b0 and b1 (****I will demonstrate this in the next class) but note the reslts

(i.e. the predicted values and the accuracy of the model)

 

If you do have 0 for male, then the predicted value for male is b0+b1*0 = b0.

4. I understood everything just fine when I went back through it after class, but Question 1 and some

of the formulas went a little fast for me at first. I'm also one of those people who panics at the sight

of long formulas (like the one for covariance), so it takes me a minute to sort through them and figure

out that they're not as complicated as they look.

 

response (T):  OK...I hope that by putting the lectures on line everyone can look at them beforehand. 

I am not sure what notation you used last semester, so I will walk throught the variance and covariance

formula again at the beginning of next class (****)

5. The review you gave of the linear regression equations confused me because it was using different

symbols than what we used in the previous course.  Also, I was not very successful in guessing where to

put the numbers for the linear reg equation (input male and female values for line) in the example/practice

question you used in class.  That makes me a little apprehensive about the course...

 

response (T): Well, we are going to start from 0 on linear regression.  Q1 in class was an attempt to persuade

you that linear regression can be seen as an algebraic tool, as we get back exactly the same results as table1.

6. Main thing was I often could not read the slides. I will move to sit closer to screen.

 

T: OK...also bring a copy of the slides...I will look at the slides from the back of the room and see what can be done (****)

 

 

Comments (0)

You don't have permission to comment on this page.