simcrv: Simulate a hospitalization curve

In jrgant/FluHospPrediction: Code for Flu Hospital Prediction Paper

Simulate a hospitalization curve

Description

Simulate a hospitalization curve

Usage

simcrv(
  print.plot = FALSE,
  print.samples = FALSE,
  print.eq = FALSE,
  verbose = FALSE,
  peakdist = dist_tf_peaks,
  hstdat = ed,
  predfits = tf_pred,
  fitseas = tf_seas,
  nu.min = 0.75,
  nu.max = 1.25,
  cv_list = cv_tf,
  lambda_type = c("lambda.min", "lambda.1se")
)

Arguments

print.plot	Logical. Print a plot of the simulated curve. Default: FALSE
print.samples	Logical. Print stochastically sampled parameters as the function runs. Default: FALSE
print.eq	Logical. Print values calculated using the modified formula from Brooks et al (see Details). Default: FALSE
verbose	Logical. Equivalent to setting both print.samples and print.eq to TRUE. Default: FALSE.
peakdist	Observed peak hospitalizations (height and week) from trend filter fits to FluSurv-NET (EIP) data.
hstdat	Observed hospitalization curves from FluSurv-NET.
predfits	Trend filter predictions based on fits to observed hospitalization curves.
fitseas	Trend filter fit objects for each observed hospitalization curve.
nu.min	Numeric. Minimum for random uniform draw governing simulated curve shifting.
nu.max	Numeric. Maximum for random uniform draw governing simulated curve shifting.
cv_list	List. containing cross-validation output for each season's trendfilter fit.
lambda_type	Character. One of "lambda.min" or "lambda.1se", each corresponding to different criteria used to select the lambda penalty for each trendfilter fit. Per the genlasso package documentation, "lambda.min" is the lambda that minimizes the cross-validated error (average error across folds), while "lambda.1se" selects the largest lambda value that produces a cross-validated error within one standard deviation of the minimum cross-validated error.

Details

Methods adapted from:

Brooks LC, Farrow DC, Hyun S, Tibshirani RJ, Rosenfeld R. Flexible Modeling of Epidemics with an Empirical Bayes Framework. PLoS Comput Biol. 2015 Aug;11(8):e1004382. Available from: http://dx.doi.org/10.1371/journal.pcbi.1004382

Value

Simulated curve depicting weekly hospitalization rates during a hypothetical flu season. Returns a list containing two nested lists, one storing sampled values from random draws, and one storing the results of feeding these values into the curve generating function.

jrgant/FluHospPrediction documentation built on May 7, 2023, 10:40 a.m.