Timeline for Is it possible to understand pareto/nbd model conceptually?
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 19, 2017 at 13:53 | comment | added | Lyuba B. | Your estimates for the Gamma parameters will depend on the data you're working with. The whole point is that the Gamma distribution can have significantly different shapes based on just two parameters, and you can let the data speak for itself, without imposing overly stringent assumptions (more or less). | |
| Apr 14, 2017 at 8:53 | comment | added | Ashok K Harnal | Thanks for such hard work and clear explanation. Reason for using Gamma distribution is as it is relatively easy to work with and is often used as a prior for positive parameters in different settings. While shapes of most gamma distribution (with different parameter values) are easy to comprehend, but to fit 'Heterogeneity in transaction rates across customers' to a gamma distribution that almost falls exponentially (for k=1, theta =2 as in Wikipedia graph [here] en.wikipedia.org/wiki/Gamma_distribution) is little difficult to comprehend. Do we exclude such behavior? | |
| Apr 14, 2017 at 8:42 | vote | accept | Ashok K Harnal | ||
| Apr 13, 2017 at 17:23 | history | answered | Lyuba B. | CC BY-SA 3.0 |