Question

@tkhunny @dan815

Answer 1

@jim_thompson5910

Answer 2

Are you able to fill out the column that has \(\LARGE \hat{y}\) in it?

Answer 3

wouldnt it be the same as the y columm?

Answer 4

no

Answer 5

then no can you explain to me

Answer 6

you take each x value and plug it into the \[\Large \hat{y} = 0.8x-0.47\] equation to generate each y-hat value

Answer 7

does that make sense?

Answer 8

yes just plug in the numbers right?

Answer 9

like for the first one is 5.13 etc

Answer 10

yep, when x = 7, the value of y-hat is 5.13

Answer 11

5.13
10.73
1.93
16.33
13.93
7.53
7.53
13.93
5.93
9.13

Answer 12

The third value of `1.93` is incorrect. Everything else is correct

Answer 13

4.33

Answer 14

yes

Answer 15

alright what next :)

Answer 16

when you get done with that column, you'll use this formula to find the error (e). Another way to state the error is to use the term residual

http://stattrek.com/regression/residual-analysis.aspx


\[\LARGE \text{residual} = (\text{observed y value}) - (\text{predicted y value})\]

\[\LARGE e = y - \hat{y}\]

Answer 17

So for example, in the first row we have

y = 6
y-hat = 5.13

which means,

\[\LARGE e = y - \hat{y}\]

\[\LARGE e = 6 - 5.13\]

\[\LARGE e = 0.87\]

the value `0.87` goes in the first row of the error column.

Answer 18

oh thsts easy

Answer 19

0.87
-0.73
-1.33
-1.33
1.07
2.47
-1.53
0.07
-0.93
0.87

Answer 20

correct

Answer 21

:D

Answer 22

for part B, you'll simply plot a bunch of points where the x coordinate is the x value from the data set. The y value will be replaced with the error e value

so the first point is (7, 0.87)
the second point is (14 , -0.73)
the third point is (6, -1.33)
etc etc

Answer 23

(c) Explain whether a linear regression line is a good fit for the data set.

Answer 24

that is the last question but the graph was not a line at all

Answer 25

look at the residual plot. Is there a pattern at all? Does any shape emerge (like a parabola)?

Answer 26

is there a way to make that bigger?

Answer 27

your residual plot should look something like this

see attached

Answer 28

then idk why mine looked like that

Answer 29

your graph is probably the same, just with a different window

Answer 30

anyways, is there a pattern at all with the residual plot?

Answer 31

no

Answer 32

In my opinion, not really. It looks like a random mess of points. This randomness means that the correlation coefficient r is very close to +1 or -1. So this is a very strong correlation.

so a linear regression is a good fit

Answer 33

oh ok since is the opposite is recommended then?

Answer 34

if you use a calculator, the correlation coefficient is r = 0.95127

which is very close to +1

Answer 35

so a linear regression line with a positive slope is a good linear fit for this (x,y) data set

Answer 36

sorry my computer froze and thank you sooo much for your help :)
Will you be online later on? Because I might have more questions

Answer 37

@jim_thompson5910

Answer 38

Yes for a little while longer

Answer 39

alright thank you :)

Answer 40

no problem