Timeline for How do you fit a Poisson distribution to table data?
Current License: CC BY-SA 3.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 22, 2016 at 8:46 | history | edited | Tim | CC BY-SA 3.0 |
added 302 characters in body
|
| Mar 22, 2016 at 8:31 | history | edited | Tim | CC BY-SA 3.0 |
added 1083 characters in body
|
| Mar 22, 2016 at 6:57 | comment | added | mavavilj | I'm still a bit unclear about the reasoning that leads to calculating the weighted mean. Why are the $y$s considered as weights? | |
| Mar 22, 2016 at 6:13 | vote | accept | mavavilj | ||
| Mar 21, 2016 at 13:51 | history | edited | Tim | CC BY-SA 3.0 |
added 2268 characters in body
|
| Mar 21, 2016 at 13:43 | comment | added | mavavilj | What's the reasonability of inferring $y$s as weights for the weighted mean (to get the MLE)? | |
| Mar 21, 2016 at 12:56 | comment | added | Tim | @mavavilj no, it means finding such value of parameters of distribution (in this case of $\lambda$) that makes the distribution function best fit the empirical distribution (see first paragraph). The last paragraph is about how your data is stored. If something is still unclear, please comment. | |
| Mar 21, 2016 at 12:53 | comment | added | mavavilj | So does fitting a model mean calculating the (empirical) $P(X=x_i)$ for each $x_i$? | |
| Mar 21, 2016 at 9:45 | history | edited | Tim | CC BY-SA 3.0 |
added 330 characters in body
|
| Mar 21, 2016 at 9:17 | history | edited | Tim | CC BY-SA 3.0 |
deleted 139 characters in body
|
| Mar 21, 2016 at 9:09 | history | edited | Tim | CC BY-SA 3.0 |
deleted 139 characters in body
|
| Mar 21, 2016 at 9:08 | comment | added | mavavilj | But how does one calculate those? I don't understand the connection between Poisson, x and y. I guess $x$ is what one plugs into the Poisson p.m.f. as $k$, but how is the Poisson p.m.f. related to the $y$s? | |
| Mar 21, 2016 at 9:02 | history | answered | Tim | CC BY-SA 3.0 |