# anyone confident in s1Watch

Announcements
Thread starter 4 years ago
#1
i just have a quick query to do with Regression.. I've done correlation so I know the concepts just bit confused..

basically i've got a mock on it tommorrow can someone explain how to get the mean of y(bar) or something along them lines.. or can someone just sumarize the chapter (send me a link) as I can't find any online
0
4 years ago
#2
(Original post by Coerce)
i just have a quick query to do with Regression.. I've done correlation so I know the concepts just bit confused..

basically i've got a mock on it tommorrow can someone explain how to get the mean of y(bar) or something along them lines.. or can someone just sumarize the chapter (send me a link) as I can't find any online
I think you mean the average of y. To find y(bar), add all the values of y, then divide by the number of values. In an exam, they will usually give y(bar) though instead of giving sample data.
0
Thread starter 4 years ago
#3
(Original post by morgan8002)
I think you mean the average of y. To find y(bar), add all the values of y, then divide by the number of values. In an exam, they will usually give y(bar) though instead of giving sample data.
How stupid of me.. ofc I ddn't mean to write it like that I know how to work a mean out. haha!

I'm just confused what is the difference between Regression and Correlation?

Except that its to do with y= a +bx and b = sxy/sxx -- which you get in the formula book?
0
4 years ago
#4
(Original post by Coerce)
How stupid of me.. ofc I ddn't mean to write it like that I know how to work a mean out. haha!

I'm just confused what is the difference between Regression and Correlation?

Except that its to do with y= a +bx and b = sxy/sxx -- which you get in the formula book?
Product moment correlation coefficient is a number between -1 and 1. The modulus(size) of the coefficient gives correlation, the sign tells you whether an increase in x causes an increase or decrease in y.

A regression line is a linear line of the form y = a + bx or y = ax + b, which allows you to estimate y for other values of x or x for other values of y.
0
Thread starter 4 years ago
#5
(Original post by morgan8002)
Product moment correlation coefficient is a number between -1 and 1. The modulus(size) of the coefficient gives correlation, the sign tells you whether an increase in x causes an increase or decrease in y.

A regression line is a linear line of the form y = a + bx or y = ax + b, which allows you to estimate y for other values of x or x for other values of y.
Oh so the aim is to basically work that out.. how do you work out A or X? I think I need to get some coffee just woken up
0
Thread starter 4 years ago
#6
Ignore what I just said.. I realized you get given a table and just work out B (sxy/sxx) and A = ybar - bxbar and im so stupid for stressing and forgetting it all sorrry
0
4 years ago
#7
(Original post by Coerce)
Oh so the aim is to basically work that out.. how do you work out A or X? I think I need to get some coffee just woken up
Ok, so in a question you will be given some data(probably x(bar) and y(bar) all the information you need anyway) and it may ask you to find the product moment correlation coefficient, so you just calculate the numerical value.

It may also ask you to find the equation of the regression line(same data) and you just find the values of a and b. It may then give you a particular value of x and ask you to estimate the value of y at this point. Just substitute the value of x into the equation you just found to find the value of y.

edit: seriously it's 9pm!
1
4 years ago
#8
all of the ax + b details can be worked out by pretty much any calculator nowadays... it is worth finding out how on yours.
0
4 years ago
#9
Another thing to note is that when using the regression line to estimate values, if it's within the range given, it is called interpolation. If it's outside of the range it is called extrapolation and is less reliable. Questions may ask you to comment on the reliability and you answer with either one of these for one mark.
0
Thread starter 4 years ago
#10
(Original post by the bear)
all of the ax + b details can be worked out by pretty much any calculator nowadays... it is worth finding out how on yours.
oh can they? I have a Casio fx-83gt plus will do some research on that haha, ty
0
Thread starter 4 years ago
#11
(Original post by morgan8002)
Another thing to note is that when using the regression line to estimate values, if it's within the range given, it is called interpolation. If it's outside of the range it is called extrapolation and is less reliable. Questions may ask you to comment on the reliability and you answer with either one of these for one mark.
Yeah thats the same concept as it was with correlation.. I just got confused with the whole Regression line cause I seen some graphs where someone predicted it and started talking about regression.
Do you get the forumlas for B = and A = or should I memorize it?
0
4 years ago
#12
(Original post by Coerce)
oh can they? I have a Casio fx-83gt plus will do some research on that haha, ty
http://cbsthurles.scoilnet.ie/blog/f...ng-a-Casio.pdf
1
4 years ago
#13
(Original post by Coerce)
Yeah thats the same concept as it was with correlation.. I just got confused with the whole Regression line cause I seen some graphs where someone predicted it and started talking about regression.
Do you get the forumlas for B = and A = or should I memorize it?
May depend on your board, but for AQA you get similar equations that if you modify slightly you get the formulae you need.
0
Thread starter 4 years ago
#14
(Original post by morgan8002)
May depend on your board, but for AQA you get similar equations that if you modify slightly you get the formulae you need.
Its not an actual paper.. just a mock on correlation and regression tommorow.. and how should i leave my last answer y = a+bx?
0
4 years ago
#15
(Original post by Coerce)
Its not an actual paper.. just a mock on correlation and regression tommorow.. and how should i leave my last answer y = a+bx?
Yes, while you're working write a = such and such
then when you've found b, b= such and such
Then write y = a + bx, with the numbers instead of a and b
0
4 years ago
#16
(Original post by Coerce)
Its not an actual paper.. just a mock on correlation and regression tommorow.. and how should i leave my last answer y = a+bx?
Also for statistics you will probably get a formula booklet in a mock(you need a formula booklet)
0
Thread starter 4 years ago
#17
(Original post by morgan8002)
Also for statistics you will probably get a formula booklet in a mock(you need a formula booklet)
Yeah I get it now.. just got confused cause of a video on examsolutions.. first time it ever confused me aha
0
X

new posts
Latest
My Feed

### Oops, nobody has postedin the last few hours.

Why not re-start the conversation?

see more

### See more of what you like onThe Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

### University open days

• Cardiff University
Wed, 27 Mar '19
• University of Portsmouth
Wed, 27 Mar '19
• Middlesex University London
Wed, 27 Mar '19

### Poll

Join the discussion

#### How old were you when you first saw porn?

I've never seen it (204)
22.54%
Before I was 12 (322)
35.58%
13 (146)
16.13%
14 (106)
11.71%
15 (56)
6.19%
16 (34)
3.76%
17 (11)
1.22%
18 (8)
0.88%
Between the ages of 19 - 25 (13)
1.44%
Over 25 (5)
0.55%