Issue #64: Vignette explaining model #405
Conversation
|
I can't figure out how to make draft PR... |
|
"Mark as draft" in bottom right. 
|
|
Ah ok thank you. |
athowes
changed the title
|
Don't forget when reading folksyou can get the HTML version in the details section of the pkgdown check |
|
Thanks @parksw3 looks great to me. I'm going to take a back seat here but I think the key questions for others are:
|
|
First quick thing: can this get added to the article here get added to the article list in |
Agree, it's an edit like this: Line 22 in cfeeadc |
|
Can someone just suggest the yaml edit to speed this up |
|
I agree with Sam that ideally we'd have some kind of extensible structure here for new models. I don't think we have that yet. At the moment things are laid out giving the required background to understand the models. Perhaps what we want to do is to move everything that goes into "the perfect model" up to background, and then have new sections for each approximation and explain at what point in the perfect model they form an approximation. We also have the At the moment I also don't think we have sign posting to many resources, aside from Park et al. and Ward et al. But I'm not sure what else would be useful. |
| &= \int_{P_L}^{P_R} \int_{S_L}^{S_R} g_P(x\,|\,P_L,P_R) f_x(y-x) \,dy\, dx | ||
| \end{aligned} | ||
| $$ | ||
| where $ g_P(x\,|\,P_L,P_R)$ represents the conditional distribution of primary event given lower $P_L$ and upper $P_R$ bounds. |
There was a problem hiding this comment.
| where $ g_P(x\,|\,P_L,P_R)$ represents the conditional distribution of primary event given lower $P_L$ and upper $P_R$ bounds. | |
| where $g_P(x \, | \, P_L, P_R)$ represents the conditional distribution of primary event given lower $P_L$ and upper $P_R$ bounds. |
|
yo yo yo what needs to happen to get this moving folks :) |
|
I can take another stab at this over the weekend & next week if someone gives more precise instructions for what needs to be done. Maybe I'll try to separate the writing into backgrounds, "perfect model", and approximations, based on @athowes's suggestions meanwhile before anyone does that (but would still appreciate instructions/suggestions)... I can also try and see if I can figure out yaml edit |
|
I've made minor modifications to text and formatting based on @athowes's earlier comments. I'm happy to do more editing/writing this week. Just let me know what needs to be done. And I think I correctly added this vignette to |
athowes
force-pushed
the
model-vignette
branch
from
041ca51 to
fb7ed4d
Compare
|
Thanks @parksw3! Have added the I've added an unpopulated section on the naive model which we should fill in. I think before merge the other sections probably need some editing for readability at a wider audience. Some of what we have at the moment is just me writing down things in quick notes. |
|
I can fill out naive model and spend some time editing for readability. Just to clarify, the naive model is the model without truncation or censoring? |
Yes that would be great, thank you!
Yes that's correct. It is just modelling a distribution on the delay directly. The code for it is here (very minimal, just basically passing We think having this model will be useful for teaching e.g. #403 and also in papers (like yours) to have something to compare the latent or marginal model to. |
Description
This draft PR will close #64.
[Describe the changes that you made in this pull request.]
Checklist