#Basic #Short #DataAnalytics post #Day2
One of the most important tools in the pocket of an data scientist specialising in Predictive Analytics is Regression. Though we mostly talk about linear and logistic regression, there are several other types of regression one needs to be aware of. First let me try and put linear and logistic regression in simple words.
#Beware this is based on my basic understanding and may have errors at certain points.
Linear Regression:
Involves predicting a variable (dependent) variable using some independent variables. Take this case of y=mx+C where y is the dependent variable and x is the independent variable
or similarly, y=ax1+bx2+cx3 and so on. Thus the value of y can be predicted given the values of x1,
x2, x3, etc. Can be performed in tools like R, using the ln() function.
Logistic Regression
Involves predicting the value of a categorical variable (recall categorical variable covered in the previous post?) using dependent variables. It can take a yes/no value or pass/fail value. For more levels in categorical variable, multinomial logistic regression maybe used. This can be done in R using the gln() function
More regression methods will be covered in future posts.
One of the most important tools in the pocket of an data scientist specialising in Predictive Analytics is Regression. Though we mostly talk about linear and logistic regression, there are several other types of regression one needs to be aware of. First let me try and put linear and logistic regression in simple words.
#Beware this is based on my basic understanding and may have errors at certain points.
Linear Regression:
Involves predicting a variable (dependent) variable using some independent variables. Take this case of y=mx+C where y is the dependent variable and x is the independent variable
or similarly, y=ax1+bx2+cx3 and so on. Thus the value of y can be predicted given the values of x1,
x2, x3, etc. Can be performed in tools like R, using the ln() function.
Logistic Regression
Involves predicting the value of a categorical variable (recall categorical variable covered in the previous post?) using dependent variables. It can take a yes/no value or pass/fail value. For more levels in categorical variable, multinomial logistic regression maybe used. This can be done in R using the gln() function
More regression methods will be covered in future posts.
Comments
Post a Comment