MINI-PROJECT 1: Finding indicators and fetching data

Introduction

This notebook covers the first mini-project, focusing on finding indicators and fetching data through the Delphi Epidata API.

Load packages

{InsightNetApr26} package ensures all required Delphi tools and their correct versions/branches are installed.

if (!requireNamespace("pak", quietly = TRUE)) install.packages("pak")
if (!requireNamespace("InsightNetApr26", quietly = TRUE)) {
    pak::pkg_install("cmu-delphi/InsightNet-apr-2026/InsightNetApr26")
}
InsightNetApr26::verify_setup()

library(epidatr)
library(epiprocess)

Finding Indicators

You can explore available indicators in the Delphi Epidata API documentation. For this example, we’ll use the “Doctor visits CLI” indicator.

Fetching Data

We’ll fetch the smoothed_adj_cli signal from the doctor-visits source for selected states over a specific time range.

res <- pub_covidcast(
    source = "doctor-visits",
    signal = "smoothed_adj_cli",
    geo_type = "state",
    time_type = "day",
    geo_values = c("mi", "ny", "tx", "pa"),
    time_values = epirange(20220101, 20220301)
)

Processing and Visualization

We convert the raw API response into an epi_df for easier analysis and plotting.

# Convert the data into epi_df format
edf <- as_epi_df(res)

# View the header of the resulting epi_df
head(edf)

An `epi_df` object, 6 x 15 with metadata:
* geo_type  = state
* time_type = day
* other_keys = signal
* as_of     = 2022-05-11
Latency (lag from as_of to latest observation by time series):
* lag  = 129–130 days
* Empty time series detected
# A tibble: 6 × 15
  geo_value signal     time_value source geo_type time_type direction issue     
  <chr>     <chr>      <date>     <chr>  <fct>    <fct>         <dbl> <date>    
1 mi        smoothed_… 2022-01-01 docto… state    day              NA 2022-03-15
2 ny        smoothed_… 2022-01-01 docto… state    day              NA 2022-03-15
3 pa        smoothed_… 2022-01-01 docto… state    day              NA 2022-03-15
4 tx        smoothed_… 2022-01-01 docto… state    day              NA 2022-03-15
5 mi        smoothed_… 2022-01-02 docto… state    day              NA 2022-03-16
6 ny        smoothed_… 2022-01-02 docto… state    day              NA 2022-03-16
# ℹ 7 more variables: lag <dbl>, missing_value <dbl>, missing_stderr <dbl>,
#   missing_sample_size <dbl>, value <dbl>, stderr <dbl>, sample_size <dbl>

# Quickly plot the data using epiprocess::autoplot
autoplot(edf, value)