Scribd
Upload
3 views

Logistic Regression

The document explains the differences between linear regression and logistic regression, highlighting that linear regression is suitable for predicting numerical responses while logistic regression is used for binary classification of categorical data. It discusses the application of logistic regression in determining probabilities for events, such as employee promotions, using a sigmoid function. A numerical example is provided to illustrate the computation of probabilities based on historical data.

Uploaded by

yogeshwar rao
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
3 views

Logistic Regression

The document explains the differences between linear regression and logistic regression, highlighting that linear regression is suitable for predicting numerical responses while logistic regression is used for binary classification of categorical data. It discusses the application of logistic regression in determining probabilities for events, such as employee promotions, using a sigmoid function. A numerical example is provided to illustrate the computation of probabilities based on historical data.

Uploaded by

yogeshwar rao
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/ 10

Difference Between Linear Regression and

Logistic Regression
• Consider the following Example:- An organization want to determine an
employee's salary hike based on their performance.
Y = Salary hike(Dependent variable)
X = Rating of employee(Independent Variable)
There is a relationship between X and Y and they are
Proportional to each other
Now lets take another
scenario
• Consider an example :- Whether The
employee should get Promotion or not
• Now there is option based answer : Yes
Or NO
Problems With linear regression
• Linear Regression predicts the numerical response but is not suitable for the
categorical values such as (Gender – Male/female, Result – Pass Or Fail, Yes
Or No, True or False)
• When categorical data is involved, we use classification.
• In binary classification we use logistic regression.
LOGISTIC REGRESSION
• When There is 2 Binary categorical data set then we use logistic regression.
• We find out the probability from 0 to 1 and then classify the data.
• 0 represents there is no possibility of an happening of an event.
• 1 represents there is a possibility of an happening of an event.
• The data set should be free of missing values.
EXAMPLE
• Consider an example :- Whether The employee should get performance or
not.
• Now we will make a Sigmoid Curve(S Curve)
• Now based on thresholds the company will decide whether the employee
should her promotion or not.
SIGMOID FUNCTION

• Sigmoid function is trying to convert the independent variable (x) into an


expression of probability which ranges from 0 to 1 with respect to the
dependent variable.

𝟏
• Formula : 𝒚 =
𝟏 𝒆 𝒙
Let's Consider an Numerical example
• The student data set has entrance marks based on the historical data of those
who are selected or not selected.
• The permeameters are given are 𝛽 = 1 𝑎𝑛𝑑 𝐵 = 8
• Assuming Marks x = 60 compute the result.
Let's Compute it
𝟏
•𝒑𝒙 = 𝟏 𝒆 𝒙
: 𝒙 = 𝜷𝟎 + 𝜷𝟏 𝒙𝟏 Here e is Euler’s Constant = 2.718

= 1 + 8*6 = 481

𝟏
𝒚= 𝟒𝟖𝟏 = 0.44
𝟏 + 𝟐 ⋅ 𝟕𝟏𝟖

You might also like