Timeline for How to build a Bayesian Model to estimate the probability distribution of the parameters given the output?
Current License: CC BY-SA 4.0
14 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 27, 2019 at 16:40 | comment | added | Sextus Empiricus | @Tbertin, you do not mind that the Bayesian way, say you use a maximum a posteriori predictor, will result in only one single outcome for each temperature. E.g. your outcome will be independent from the particular $\beta$ and $\alpha$ that may generate the temperature, and you will always end up selecting the $\beta$ and $\alpha$ on the iso-temperature line for which the prior joint probability $P(T=x,\alpha=y,\beta=z) = P(\alpha=x,\beta=y) \delta(T-x)$ is maximum. That is, you will be just picking the maximum posterior $ P(\alpha,\beta)$ on the isoline for the T that you measure. | |
| Jan 24, 2019 at 21:26 | comment | added | Tbertin | yes, it is my point | |
| Jan 23, 2019 at 8:03 | comment | added | Sextus Empiricus | I notice that some people have given an answer in terms of tackling the problem in Bayesian way. Computing some posterior $P(\alpha,\beta | T)$ based on a prior $P (\alpha,\beta)$ and an observed $T$. Is that what you are looking for? | |
| Jan 22, 2019 at 15:30 | vote | accept | Tbertin | ||
| Jan 19, 2019 at 14:35 | comment | added | Tbertin | i have only one estimation per couple | |
| Jan 19, 2019 at 13:31 | answer | added | Salma Bouzid | timeline score: 1 | |
| Jan 19, 2019 at 10:09 | answer | added | Sextus Empiricus | timeline score: 2 | |
| Jan 19, 2019 at 7:22 | comment | added | Karel Macek | Just curious:how many experiments or records do you have for one combination of $\alpha$ and $\gamma$? | |
| Jan 19, 2019 at 0:00 | answer | added | Dave Harris | timeline score: 1 | |
| Jan 18, 2019 at 20:55 | history | edited | Tbertin | CC BY-SA 4.0 |
added 470 characters in body
|
| Jan 18, 2019 at 15:28 | comment | added | Tbertin | Hi, thank you for your answer, but as you can see, a lot of cells have the same temperature, it means that is for a single temperature, you have a bunch of possible alpha and beta... the aim is to find their probability distribution... | |
| Jan 17, 2019 at 21:53 | comment | added | Karel Macek | I suppose you have a matrix that you use for the visualization. You can find the cell in that matrix that is closest to 33C. Would this work? | |
| Jan 17, 2019 at 21:25 | answer | added | benso8 | timeline score: 1 | |
| Jan 17, 2019 at 21:12 | history | asked | Tbertin | CC BY-SA 4.0 |