Comments (3)
Probably a good idea to explain what is meant by day of reference vs report in this document too.
Agree. We need to keep it concise though to maintain this as a quick start vs a longer intro to the package.
What is (1 | day)? Is this the random walk?
Its a random intercept by day,
"Rather than using the methods supplied for epinowcast directly, package functions can also be used to extract nowcast posterior samples, summarise them, and then plot them" Is this sentence missing something? I'm confused what "package functions" are referring to. It sounds like you might be referring to a different package? Not sure why functions vs methods are different here...
This is complete but might need an edit. The distinction this is drawing is that there are user facing S3 methods for the output for epinowcast
(i.e. methods) and then normal functions that underlie these methods that more advanced users can use if they wish.
Am curious why there is a consistent underestimation. Would be useful to give some information (e.g., what things you can try, what to read) for readers who want to learn more (but don't have to show how to actually do more complicated things)...
Yes, this is a good point and I like the idea of pointing users towards other resources for learning more/ideas of things to try. Some of the obvious things to try (which are tried in the german vignette are time-varying reporting delay distributions and day-of-the-week effects for the reporting delay as well as for the date of report. Once we have them implemented non-parametric model specifications will likely do well here. Lastly, there is a clear day of the week signal in the final reported counts this could be included in the expectation model as a random effect on the day_of_week
. A more mechanistic expectation model might also help (i.e one with some kind of growth rate vs a random walk on the log scale).
from epinowcast.
Fancy making some of these changes @parksw3 (or any others for the comments in the related issues).
from epinowcast.
This seems like we should address it for 0.2.1
. Currently no one is assigned but ideally we would be aiming for a release candidate tomorrow. If anyone can turn it around that quick it would be great but if not I will address tomorrow morning and then ping @parksw3 to see if it has covered his comments.,
from epinowcast.
Related Issues (20)
- Hex sticker logo has old URL in HOT 1
- Package experience for new user? HOT 2
- Add/Check support for using a `Cmdstan` fit as initial conditions for a new fit + support passing from pathfinder to NUTS
- Map parameter model names to design matrix names + split based on model concepts
- Improve documentation of the formula interface
- Better link model modules with model equation writeup
- Partial match of various elements in `prior` object HOT 2
- `enw_nowcast_samples()` and `enw_nowcast_summary()` give unexpected output when not working with daily time steps HOT 2
- Fix Rmd check on main HOT 1
- Implement autotest to test function inputs
- Support for fixed reporting schedule via delay probability aggregation HOT 1
- Add support for aggregation of probabilities in `expected_obs` HOT 1
- Design `structural` input to `enw_report()` for aggregating probabilties HOT 2
- Design input to the Stan code that would link user-supplied input to the aggregation argument to `expected_obs` HOT 2
- Modify `enw_report()` function documentation and add an example to `epinowcast` documentation
- Add unit/integration tests for aggregating probabilties functionality
- Update to use current function for exposing stan functions
- Explore methods with potentially better computational performance for probability aggregation in `expected_obs()` HOT 1
- Error occurs when fitting the model in: Estimating the effective reproduction number in real-time for a single timeseries with reporting delays HOT 1
- Add render-readme like actions for compiling precompiled vignettes
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from epinowcast.