Notes For Linear Regression and Correlation CA Foundation Paper – 3

Linear Regression and Correlation : CA Foundation Paper - 3
Linear Regression and Correlation

Notes For Linear Regression and Correlation CA Foundation Paper – 3

Correlation, in the account and venture enterprises, is a measurement that gauges how much two protections move according to one another. Relationships are utilized in the cutting edge portfolio of the executives, figured as the connection coefficient, which has a worth that must fall between – 1.0 and +1.0.


Correlation is a measurement that gauges how much two factors move according to one another. In account, the correlation can gauge the development of stock with that of a benchmark file, for example, the S&P 500. Correlation estimates affiliation yet doesn’t show if x causes y or the other way around, or if the affiliation is brought about by a third–maybe inconspicuous factor.


A correlation or simple analysis of linear regression may ascertain whether there is a significant correlation between the two numeric variables. An analysis of correlation provides information on the intensity and direction of the linear relationship between two variables, while a simple analysis of linear regression estimates parameters in a linear equation that can be used to predict values depending on the other for one variable.


An ideal positive correlation implies that the relationship coefficient is actually 1. This suggests that as one security moves, either up or down, the other security moves in lockstep, in a similar course. An ideal negative correlation implies that two resources move in inverse ways, while a zero correlation suggests no straight relationship by any means.


For instance, enormous cap shared assets by and large have a high sure correlation to the Standard and Poor’s (S&P) 500 Index or almost one. Little cap stocks have a positive relationship to the S&P, yet it’s not as high or around 0.8. In any case, put alternative costs and their fundamental stock costs will in general have a negative correlation. For audit, a put alternative gives the proprietor the right, yet not the commitment, to sell a particular measure of basic security at a pre-decided cost inside a predetermined time period. Put alternative agreements become more beneficial when the fundamental stock value diminishes. At the end of the day, as the stock cost expands, the put alternative costs go down, which is an immediate and high-extent negative relationship.

The Formula for Correlation is as follows:

Linear Regression and Correlation

Linear Regression and Correlation

Browse the video lecture Correlation and Regression of CA Foundation/CMA Foundation

Linear Regression

A linear regression examination produces gauges for the slant and capture of the direct condition anticipating a result variable, Y, in light of estimations of an indicator variable, X. An overall type of this condition is demonstrated as follows

Y = b 0 + b1 . X

The intercept, b 0, is the anticipated estimation of  Y when X = 0.

. The incline,b1 , is the average change in Y for each one-unit increment in X. This shows the heading of the linear relationship among X and Y, the slant gauge permits an understanding for how Y changes when X increments. This condition can likewise be utilized to foresee estimations of Y for estimation of X.

Linear Regression and Correlation


Inferential tests can be run on both the relationships and incline gauges determined from an arbitrary example from a populace. The two investigations are t-tests run on the invalid theory that the two factors are not directly related. Whenever we run on similar information, a connection test and an incline test give a similar test measurement and p-value.


1. Random samples

2. Independent observations

3. The predictor variable and outcome variable are linearly related (assessed by visually checking a scatterplot).

4. The population of values for the outcome are normally distributed for each value of the predictor (assessed by confirming the normality of the residuals).

5. The variance of the distribution of the outcome is the same for all values of the predictor (assessed by visually checking a residual plot for a funneling pattern).



Regression Example

The quality of UV beams fluctuates by scope. The higher the scope, the less presentation to the sun, which compares to a lower skin malignant growth hazard. So, where you live can affect your skin malignant growth hazard. Two factors, malignant growth death rate and scope were gone into Prism’s XY table. The Prism diagram (right) shows the connection between skin malignant growth death rate (Y) and scope at the focal point of a state (X). It bodes well to process the connection between’s these factors, yet making it a stride further, how about we play out a regression examination and get a prescient condition.

The connection among X and Y is summed up by the fitted regression line on the chart with a condition: death rate = 389.2 – 5.98*latitude. In view of the slant of – 5.98, every 1-degree increment in scope diminishes passings because of skin malignant growth by roughly 6 for each 10 million individuals. Since regression investigation creates a condition, in contrast to connection, it tends to be utilized for expectation.


For instance, a city at scope 40 would be required to have 389.2 – 5.98*40 = 150 passings for every 10 million because of skin malignant growth every year. Regression additionally takes into consideration the translation of the model coefficients:

1. Slope: each one-degree increment in scope diminishes mortality by 5.98 passings per 10 million.

2. Intercept: at 0 degrees scope (Equator), the model predicts 389.2 passings per 10 million. In spite of the fact that, since there is no information at the block, this expectation depends intensely on the relationship keeping up its straight structure to 0.


Key similarities between correlation and linear regression


1. Both measure the heading and quality of the connection between two numeric factors.

2. At the point when the correlation (r) is negative, the regression slope (b) will be negative.

3. At the point when the correlation is positive, the regression slope will be positive.

4. The correlation squared (r 2 )has exceptional importance in basic direct relapse. It speaks to the extent of variety in Y clarified by X.


Key differences between correlation and linear regression

1. Regression endeavors to set up how X makes Y change and the aftereffects of the examination will change if X and Y are traded. With connection, the X and Y factors are compatible.

2. Regression accepts X is fixed with no blunder, for example, a portion sum or temperature setting. With a relationship, X and Y are normally both irregular variables*, for example, stature and weight or circulatory strain and pulse.

3. Correlation is a solitary measurement, though relapse creates a whole condition.

Must read – Indian Contract Act, 1872


Prepare now for a  CA Foundation Exams with our CA Foundation Online Classes right from your home.

Here, at Takshila Learning, we offer CA Foundation Video Lectures and CA Foundation Pen drive Classes to help you in CA Exam Preparation. And we also offer Online Tuition Classes from Preschool to Class 12, Online Classes for Professional Courses, i.e., CA Course/ CS Online Classes / CMA Online Classes / ACCA IFRS, Skill Development courses, i.e., Python for EDA Course, Digital Marketing Course, Competitive Exams, i.e., SSC / Banking online classes / Railways.

Register now to enquire more!


Call at 8800999280 / 8800999283 / 8800999284 fill the form for any other details:


Share and Enjoy !

0 0

0 responses on "Notes For Linear Regression and Correlation CA Foundation Paper - 3"

Leave a Message

Your email address will not be published. Required fields are marked *

© 2021-22 Takshila Learning. All Rights Reserved.
Request Callback
close slider
For course & fee related queries, Leave your details and our counsellor will get back to you or Call us at 8800-999-280
  • This field is for validation purposes and should be left unchanged.