lifetimes
lifetimes copied to clipboard
Lifetime value in Python
I can't find any info on this parameter in the `plot_cumulative_transactions` function. According to the documentation, `t_cal` is "A marker used to indicate where the vertical line for plotting should...
How to deal with conditional_expected_number_of_purchases_up_to_time() returning 0 expected transactions even in a prolonged holdout time period for yearly customers? I have a dataset which consists of weekly, monthly, as well...
Extend package with functionality described in [this paper](http://brucehardie.com/notes/019/time_invariant_covariates.pdf): > Bruce Hardie, *Incorporating Time-Invariant Covariates into the Pareto/NBD and BG/NBD Models*.
Hi! I am trying to fit Beta-Geometric/Beta-Binomial Model model but got a lot of errors like invalid value, overflow or not converge. I am thinking maybe I am using the...
I'm running into a situation where the results of my BGF probability score defaults to 1 for users that have no frequency (one-time purchasers). Like most companies, I have customers...
I'm working with a package-data: packages last a month so most users would make a purchase once a month. When using summary_data_from_transaction_data would it be better to use freq='M' or...
1. I am aware to the fact that the Gamma-Gamma Model ignores the monetary value of first purchase, and only data with frequency>0 should be provided to the model. Based...
Hi everyone, Practical theory question for you all :) Bruce Hardie's paper on the Gamma-Gamma clearly states that spend data is usually right-skewed, and my dataset is no different -...
Hello, It seems that I found the inconsistence meanings when I use `calibration_and_holdout_data` function. My transaction data have 3 fields, `customer_id`, `finish_time`, `monetary value`. Then I use `calibration_and_holdout_data` function ```python...
What are the best approaches to include covariate in the number of transaction and/or monetary value predictions? Are are there any plans to extend this package to incorporate such models?...