dadnow - Nowcasting Discharge Abstract Database (DAD)

Example Usage

Start by loading the data. David has provided functions to produce sample data in the format expected by the functions.

First, the DAD data:

library(dadnow)

set.seed(2112)
data.dad  = fake_dad()
data.expl = fake_expl(dad = data.dad)

head(data.dad)

        date virus geo count percapita
1 2020-01-01   VXX GZZ    57  0.000057
2 2020-01-08   VXX GZZ    99  0.000099
3 2020-01-15   VXX GZZ   141  0.000141
4 2020-01-22   VXX GZZ   184  0.000184
5 2020-01-29   VXX GZZ   254  0.000254
6 2020-02-05   VXX GZZ   300  0.000300

Next, generate the explanatory variables

head(data.expl)

        date virus geo thecount
1 2020-01-01   VXX GZZ       31
2 2020-01-08   VXX GZZ       49
3 2020-01-15   VXX GZZ       66
4 2020-01-22   VXX GZZ       98
5 2020-01-29   VXX GZZ      137
6 2020-02-05   VXX GZZ      133

prm = list(
  family = 'lm',
  data.dad = data.dad,
  data.expl = data.expl,
  varname.dad = 'count',
  varname.expl = 'thecount',
  date.range.fit = c('2020-01-10', '2020-12-15')
)

fitted.model = fit_model(prm)

lapply(fitted.model, class)

$model
[1] "lm"

$data
[1] "data.frame"

$varname.dad
[1] "character"

$varname.expl
[1] "character"

$family
[1] "character"

$date.range.fit
[1] "Date"

g.fit = plot_fitted_model(fitted.model)
g.fit$xy

g.fit$timeseries

n = 15
newdata.expl = data.frame(
  date = data.expl$date[1:n] + 360,
  virus = 'VXX',
  geo = 'GZZ',
  thecount = rpois(n, lambda = 20*c(1:n))
)

a = nowcast(fitted.model = fitted.model,
            newdata.expl = newdata.expl)
g.nowc = plot_nowcast(nowc = a)
g.nowc