Advanced Data Analytics
Added on 2023-06-03
7 Pages1235 Words218 Views
|
|
|
Running head: ADVANCED DATA ANALYTICS
Advanced Data Analytics
Name of the Student
Name of the University
Course ID
Advanced Data Analytics
Name of the Student
Name of the University
Course ID
1ADVANCED DATA ANALYTICS
Table of Contents
Question 3........................................................................................................................................2
Logistic model.............................................................................................................................2
Decision tree................................................................................................................................4
References list..................................................................................................................................6
Table of Contents
Question 3........................................................................................................................................2
Logistic model.............................................................................................................................2
Decision tree................................................................................................................................4
References list..................................................................................................................................6
2ADVANCED DATA ANALYTICS
Question 3
In today’s world, social media plays an important role in business promotion, launching
and marketing of products. Larger views and sharing advertisement has significant influence on
marketing of a product. In the given scenario, the marketing or advertising companies are
interested to know whether a Twitter messages will spread as a meme or not. Advertising or
marketing companies always want the twitter message to be spread.
Statistical models are used to predict outcome of an events. The simplest and most
commonly used estimation technique is ordinary least square. The problem here is that the
outcome of the stated event is binary (Discacciati, Crippa and Orsini 2017). The variable
modeling whether a twitter message is spread or not is a binary variable with 1 denoting that the
message will spread and 0 denotes that the message will not spread. In case of binary dependent
variable, the widely used method of ordinary least square fails to provide best linear unbiased
estimator (BLUE). OLS thus produces a biased and inefficient outcome. Alternative methods
thus needs to be used for predicting binary outcomes. Models for such binary outcomes are
difficult compared to simple linear regression model. These models are generally complicated to
develop and difficult to interpret (Hayes and Montoya 2017). Because of discrete nature of the
outcome, it is hard to predict whether a tweet will go viral not.
Logistic model
One approach to predict whether a tweet will go viral or not is logistic regression model.
The logistic regression model is applicable to find out the probability of success and failure of an
event. Logistic model should be used in cases where the targeted dependent variable contain two
Question 3
In today’s world, social media plays an important role in business promotion, launching
and marketing of products. Larger views and sharing advertisement has significant influence on
marketing of a product. In the given scenario, the marketing or advertising companies are
interested to know whether a Twitter messages will spread as a meme or not. Advertising or
marketing companies always want the twitter message to be spread.
Statistical models are used to predict outcome of an events. The simplest and most
commonly used estimation technique is ordinary least square. The problem here is that the
outcome of the stated event is binary (Discacciati, Crippa and Orsini 2017). The variable
modeling whether a twitter message is spread or not is a binary variable with 1 denoting that the
message will spread and 0 denotes that the message will not spread. In case of binary dependent
variable, the widely used method of ordinary least square fails to provide best linear unbiased
estimator (BLUE). OLS thus produces a biased and inefficient outcome. Alternative methods
thus needs to be used for predicting binary outcomes. Models for such binary outcomes are
difficult compared to simple linear regression model. These models are generally complicated to
develop and difficult to interpret (Hayes and Montoya 2017). Because of discrete nature of the
outcome, it is hard to predict whether a tweet will go viral not.
Logistic model
One approach to predict whether a tweet will go viral or not is logistic regression model.
The logistic regression model is applicable to find out the probability of success and failure of an
event. Logistic model should be used in cases where the targeted dependent variable contain two
End of preview
Want to access all the pages? Upload your documents or become a member.
Related Documents