enter image description here

I confused about decision tree such as this picture why repeat decision node.Could you please explain that decision tree. thank you

  • $\begingroup$ On this picture, the variable range was split into three subranges. $\endgroup$ – ttnphns Nov 16 '18 at 11:55

The proportion of "open" responses apparently varies across the variable "added_date_m". In particular, the proportion appears to be extremely close to 100% if added_date_m is larger than 49. Moreover, if added_date_m is lower or equal 49 but larger than 26, the proportion is still high but not quite as close to 100%.

More generally, if there is a smooth or even linear relationship between the mean of the response variable and the regressor variable, then the tree will require repeated splits in the same variable to approximate this.

  • 1
    $\begingroup$ If this solved your problem, please accept the answer here on StatsExchange so that it also is flagged as solved. $\endgroup$ – Achim Zeileis Nov 19 '18 at 20:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.