Forums › Ask ACCA Tutor Forums › Ask the Tutor ACCA PM Exams › Regression analysis

- This topic has 5 replies, 2 voices, and was last updated 1 year ago by John Moffat.

- AuthorPosts
- December 18, 2021 at 6:51 pm #644528
The formula sheet of paper PM has been updated and now it includes Regression analysis formulas. True?

I have not seen your lecture yet but please state that there are formulas for regression analysis but no formula for time series analysis. True?

Thank you for your time :))

December 19, 2021 at 6:34 am #644535True for both 🙂

December 19, 2021 at 1:02 pm #644546I have seen your lecture on regression analysis but I could not understand it completely.

What I understand is that regression analysis is used to identify a straight line that shows a linear relationship between two variables.

Is it true that these two variables are total cost (y) and the number of units (x)?

It shows the relationship between two variables that if one variable (number of units (x)) is increasing then the other variable (total costs (y)) will also be increasing. It simply means that with more units produced there will be more total costs of production.

To show this relationship we need to calculate the fixed cost and variable cost using regression analysis formula

Y = a + bx

We could be asked regression analysis calculations only in sections A and B in paper PM?

December 19, 2021 at 2:58 pm #644550Your understanding is correct except for the fact that it can be used to express the relationship of any two variables – it doesn’t have to be the total cost and the number of units (although that is the most likely relationship to exist in Paper PM).

It can be asked in any section of the exam.

December 19, 2021 at 3:38 pm #644552Thank you so much. 🙂

Please say about this too.Correlation Coefficient:

It measures how linear the relationship is between two variables such as whether it is perfectly positive linear correlation or not using this formula.r = (n(Exy) – Ex Ey / squareroot n(Ex^2 – (Ex)^2)) x (nEy^2 – (Ey)^2)

+ 1 = The points are exactly on a straight-line going upwards

1 = The points are closely on a straight-line going upwards

– 1 = The points are exactly on a straight-line going downwards

0 = No correlation at all (straight-line is not possible to draw)The correlation coefficient is used only to see whether the straight-line which is passing through the points are perfectly or not perfectly positive or negative.

Is it correct?

Is it also correct that linear relationship refers to the straight-line which passes through points (or variable points) but the linear relationship refers to only two variables only. I don’t understand why is it important though?

December 20, 2021 at 6:56 am #644570The correlation coefficient is used to see how near (or how far) away the points are from being in a straight line.

In practice several variables can affect each other – for instance being ill can be cause by lots of different things. Regression only looks at the relationship between two variables.

- AuthorPosts

- You must be logged in to reply to this topic.