# Subscription Prediction
[2020-03-13]
* prediction of transactions (remove ISP and auto-subscription users)
* xgboost model
* period A (12 weeks)
* 
* period B (12 weeks)
* 
* pearson correlation with transactions (remove ISP and auto-subscription users) (2010~2019)

* pearson correlation with transactions (remove ISP and auto-subscription users) (2017~2019)

* pearson correlation with transactions (remove ISP and auto-subscription users) (2019~)

[2020-03-11]
* Downloads with spotify



* trans stats
http://redash-rdc.kkinternal.com/queries/1898/source#4263
* new user trans stats
http://redash-rdc.kkinternal.com/queries/1899/source#4266
* 30-days one time trans conversion

[2020-02-26]
* monthly conversion -> weekly transaction
* weekly data: https://docs.google.com/spreadsheets/d/1NPcUtujr1MGQTy4wBUvqenWJjJFm0tBDibN-LR171zk/edit?usp=sharing
* transaction total prediction (with AR model)
* 
* split with user type
* AR
* old user (registered more than 3 months)
* 
* new user (registered less than 3 months)
* 
* STL decomposition
* old user (registered more than 3 months)
* 
* new user (registered less than 3 months)
* 
* ACF & PACF
* old user (registered more than 3 months)
* 
* new user (registered less than 3 months)
* 
* SARIMAX
* old user (registered more than 3 months)
* 
* new user (registered less than 3 months)
* 
* VAR (variables: register, play count, download)
* new user (registered less than 3 months)
* 