Hi Professor, I have a question about the residual plot, like you said, the prediction is the probability and our response variable is 0 or 1. But why some residuals are greater or smaller than 1 or -1
Thanks, the lecture was very informative. I wanted to know whether the results will have significance in logistic regression if remove out non-significant/not correlated variables applying the linear regression using the 5% significance test.
Hello sir, Thanks for your great lectures, it was really helpful and useful for me. And I would like to ask could you deliver a lecture on “simulation based on Logistic regression results”?
Hey @aleksandra you can only perform cor() function on only numeric variables, if your categorical data is A,B,C,D it won't work unless the variable is numeric.
The variable Sexmale is named that way because it takes the value 1 for male and 0 for female. Thus, this variable encodes the gender, both male and female. This variable is significant, which means gender is significant. I could have named the variable "gender", but then you would have to remember whether 1 meant male or female.
Nice information Sir. I got some knowledge by watching this video.
great video ! super helpful and to the point !
Very nice job done, Thank you very much Professor!
Thank you so much for your effective video sir.
Great Video, thanks professor!!! I learnt a lot~!
Thanks Chris...
very good job, thanks!
Hi Professor, I have a question about the residual plot, like you said, the prediction is the probability and our response variable is 0 or 1. But why some residuals are greater or smaller than 1 or -1
Thank you sir.. bookmarked your website
Thanks, the lecture was very informative. I wanted to know whether the results will have significance in logistic regression if remove out non-significant/not correlated variables applying the linear regression using the 5% significance test.
Is there a video where you discuss null deviance and the other associated output from the summary?
Hello sir,
Thanks for your great lectures, it was really helpful and useful for me. And I would like to ask could you deliver a lecture on “simulation based on Logistic regression results”?
getting an error
Error in dimnames(X) 0], use.names = FALSE)) :
length of 'dimnames' [2] not equal to array extent
help
Hi Prof, I was wondering what the format(class1_survival_rate, digits=3) mean? in the survival rate code.
It just sets the number of digits used to display the numerical result.
I love you !
Can we really use cor() for categorical data?
Thank you in advance
Hey @aleksandra you can only perform cor() function on only numeric variables, if your categorical data is A,B,C,D it won't work unless the variable is numeric.
ok. at 28:50 Sex male is Significant.. what happens to our reference level Sex female? is it significant or neutral?
The variable Sexmale is named that way because it takes the value 1 for male and 0 for female. Thus, this variable encodes the gender, both male and female. This variable is significant, which means gender is significant. I could have named the variable "gender", but then you would have to remember whether 1 meant male or female.
how to test that model with test data .....i had build one model but idont know how to test my model with training dataset....can you help me out
Check out Lecture 62: Building Models.
Use some spaces if there is bulk of code exist , its easy to read for who is learning ..by the way video is good
specifically talking about line 31 to 43...
take this feedback in positive way.. :)
hey, why this video is not clear.