Add forecasting vignette #6

EmilieFinch · 2024-08-30T11:46:47Z

This PR adds a vignette outlining an example forecasting workflow using relative R0 values estimated with the temperature_r0 function within a gam forecasting framework. This is exploratory work, and the vignette documents current limitations and ideas for potential future directions.

This PR also updates the fiji_2014 dataset to include population size, needed for forecasting.

github-actions · 2024-08-30T11:48:40Z

This pull request:

Adds 4 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

This uses TSIR-like forward propagation

New example to be incorporated

Also convert rR0 to log(rR0) in GLM for consistency

Use offset for R0 and infections, allow coefficient to represent S^n

Data not yet in package

github-actions · 2024-12-03T13:12:12Z

This pull request:

Adds 4 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

github-actions · 2024-12-03T13:17:22Z

This pull request:

Adds 5 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

github-actions · 2024-12-05T10:58:53Z

This pull request:

Adds 5 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

github-actions · 2024-12-05T11:02:27Z

This pull request:

Adds 5 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

adamkucharski

Thanks for sharing. I'm getting an error in gam-forecasting-workflow.Rmd but otherwise looks good. Will also look at tidying up linting in gam-simulation-workflow.Rmd before merge.

adamkucharski · 2024-12-05T11:00:17Z

vignettes/gam-forecasting-workflow.Rmd

+
+```{r}
+
+all_forecasts <- NULL 


I'm getting this error for this chunk:

Generating forecasts for prediction week starting: 2014-01-06 Warning: basis dimension is larger than number of unique covariatesError in select(train_data, date, date_numeric, rR0, log_pop_susceptible, : unused arguments (date, date_numeric, rR0, log_pop_susceptible, log_weighted_lagged_cases, cases)

adamkucharski · 2024-12-05T11:00:32Z

vignettes/gam-forecasting-workflow.Rmd

+Now we wrangle the Fiji 2014 data for forecasting and set our forecast horizon (how many weeks ahead we want to forecast) and the dates we would like our forecasts to start from.
+
+```{r}
+model_data <- fiji_2014 |> 


Need to define pop? Have pushed an edit

Added dplyr::select avoid namespace clash with dplyr and MASS… Also edited date matching that didn’t seem to be working in `weekly_forecasts[[match(prediction_start,prediction_dates)]] `

github-actions · 2024-12-05T15:46:27Z

This pull request:

Adds 5 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

Fix S and allow flexibility on lagged cases

github-actions · 2024-12-06T11:30:48Z

This pull request:

Adds 5 new dependencies (direct and indirect)
Adds 1 new system dependencies
Removes 0 existing dependencies (direct and indirect)
Removes 0 existing system dependencies

Reach out on slack (#code-review or #help channels) to double check if there are base R alternatives to the new dependencies.

(Note that results may be inaccurate if you branched from an outdated version of the target branch.)

EmilieFinch added 3 commits August 30, 2024 12:39

Update fiji example data to include population size

33cf1d8

Add vignette for example forecasting workflow

444b2e2

Update data documentation

24003b7

EmilieFinch requested review from adamkucharski and joshwlambert August 30, 2024 11:46

adamkucharski and others added 16 commits November 4, 2024 21:19

Add step-by-step vignettes

80a633c

This uses TSIR-like forward propagation

Add suggestion

75ebf36

Add bootstrap uncertainty

0ae657e

New example to be incorporated

Tidy functions

8b0cd1a

Tidy functions and plots

cfd5996

Swap back to Fiji scenario

75c9a0a

Add 2017 example

adc4bd8

Also convert rR0 to log(rR0) in GLM for consistency

Update forecast model

5975023

Use offset for R0 and infections, allow coefficient to represent S^n

Add R0 to plots

581fb8e

Tidy code for knitting

c5042da

Remove knitting for 2017 outbreak

f5733fe

Data not yet in package

Tidy comments

abc5ba1

Tidy Rmd with Fiji example

e021e52

Fix missing variable

cd9f3f7

Fix typo and add date origin

d45740a

Merge branch 'main' into forecasting-vignette

150bbb5

Add abind as import

8748996

Fix index entry titles

d548466

Define pop in forecasting script

50eeac0

adamkucharski reviewed Dec 5, 2024

View reviewed changes

Name clash and matching edit

9fa8d09

Added dplyr::select avoid namespace clash with dplyr and MASS… Also edited date matching that didn’t seem to be working in `weekly_forecasts[[match(prediction_start,prediction_dates)]] `

Edit offset

f5fb915

Fix S and allow flexibility on lagged cases

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add forecasting vignette #6

Add forecasting vignette #6

EmilieFinch commented Aug 30, 2024

github-actions bot commented Aug 30, 2024

github-actions bot commented Dec 3, 2024

github-actions bot commented Dec 3, 2024

github-actions bot commented Dec 5, 2024

github-actions bot commented Dec 5, 2024

adamkucharski left a comment

adamkucharski Dec 5, 2024

adamkucharski Dec 5, 2024

github-actions bot commented Dec 5, 2024

github-actions bot commented Dec 6, 2024


		```{r}

		all_forecasts <- NULL

Add forecasting vignette #6

Are you sure you want to change the base?

Add forecasting vignette #6

Conversation

EmilieFinch commented Aug 30, 2024

github-actions bot commented Aug 30, 2024

github-actions bot commented Dec 3, 2024

github-actions bot commented Dec 3, 2024

github-actions bot commented Dec 5, 2024

github-actions bot commented Dec 5, 2024

adamkucharski left a comment

Choose a reason for hiding this comment

adamkucharski Dec 5, 2024

Choose a reason for hiding this comment

adamkucharski Dec 5, 2024

Choose a reason for hiding this comment

github-actions bot commented Dec 5, 2024

github-actions bot commented Dec 6, 2024