Scribd
Upload
12 views4 pages

B.Tech - 5thsem - KCS055 - Unit 2 - 1

Uploaded by

akg.uk14
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
12 views4 pages

B.Tech - 5thsem - KCS055 - Unit 2 - 1

Uploaded by

akg.uk14
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Regression Analysis is a form of predictive modelling techniques which investigates the relationship

between a dependent and independent variable.

1. It is a statistical tool used to understand and quantify the relation between two or more
variables.

2. Regression range from simple model to highly complex equations

3. Two primary uses of regression in business are forecasting and optimization.

4. 4. In addition to helping managers predict such things as future demand for their products ,
regression analysis helps fine tune the manufacturing and delivery process

Objectives:

Establish if there is a relationship between two variables .

a. More specifically , establish if there is a statistically significant relationship between the two .

b. Examples: Income and spending ,wage and gender,student height

Objectives:

Forecast new observations:

a. Can we use what we know about the relationship to forecast unobserved values?

b. Examples : What will are sales be over the next quarter?What will the ROI of a new store
opening be contigent on store attributes

Variable’s Role:

Dependent Variable:

1. This is the variable whose values we want to explain or forecast.

2. Its values depend on something else.

3. We denote it as Y.
Variable’s Role:

Independent Variable:

1. This is the variable that explains the other ones.


2. Its values are independent.
3. We denote it as X.

The Magic: A linear equation

You may remember one of these:

y=a+bx

Y=mx+b

In the stats world, we just use a different notation:

Y=beta0+betax

We call it “linear” because the equation represents a straight line in a bi-dimensional plot.

Simple Linear Regression Model:

Y=beta0+beta1x+epsilon

y= dependent variable

X= independent variable

Beta0= constant or intercept

Beta1=x slope or coefficient

Epsilon= error term

Three major uses of regression analysis:

1. Determining the strength of predictors


2. Forecasting an effect

3. Trend Forecasting

Core Concept: In linear regression data is modelled using a straight line but in logistic regression
probability of some obtained event is represented as a linear function of a combination of
predictor variables.

Used with: In linear regression we use continuous variable but in logistic we use categorical
variables.

Output/Prediction: in case of linear we get value of the variable but in logistic we get probability
of occurrence of event.

Accuracy and Goodness of fit: In case of linear regression ,it is measured by loss, R squared,
Adjusted R squared etc but in logistic regression it is measured by Accuracy, Precision , Recall, F1
score, ROC curve, confusion matrix.

Selection Criteria:

1. Classification and regression capabilities

2. Data Quality

3. Computational Complexity

4. Comprehensible and transparent

Where it is used:

1. Evaluating trends and Sales Estimates

2. Analyzing the impact of Price Changes

3. Assessment of risk in financial services and insurance domain.


What is R-square:

1. R-squared value is a statistical measure of how close the data are to the fitted regression line.

2. It is also known as coefficient of determination , or the coefficient of multiple determination.

Drawbacks:

1. Need to evaluate the whole dataset every time.

2. Computation cost.

3. Memory requirement increases as dataset increases.

Drawbacks:

1. One of the problems with linear regression is that it tries to fit a constant line to your data
once the model was created

2. Such behavior might be okay when your data follows linear pattern and does not have much
noise .

3. However, when dataset is not linear ,linear regression tends to under fit the training data.

You might also like