Generate insecticide resistance trend scenarios — run_resistance

Extends mortality time series beyond a specified year using one of three projection methods: constant trend (flat-line), continued trend (model extrapolation), or hybrid trend (average of both).

Supports flexible grouping and can handle any IR metric column name.

Usage

run_resistance_trend(
  df_IR,
  year_cut = 2025,
  n_years_to_add = 5,
  continue_trend_start_year = 2010,
  model_type = "unchanged",
  trendmethod = c("constantTrend", "continuedTrend", "hybridTrend"),
  group_cols = c("adm0", "adm1"),
  value_col = "mean_ir"
)

Arguments

df_IR: Data frame containing insecticide resistance data with grouping columns, year column, and value column.
year_cut: Numeric. Last year of observed data. Projections start from year_cut + 1. Default is 2025.
n_years_to_add: Numeric. Number of years to project forward. Default is 5.
continue_trend_start_year: Numeric. First year to include in model fitting for continued trend. Default is 2010.
model_type: Character scalar. Regression family for continued trend. One of "gaussian", "binomial", "quasibinomial", "lm", "gam", or "unchanged". If "unchanged", uses existing data values directly without refitting. Default is "unchanged".
trendmethod: Character scalar. Projection method: "constantTrend" (flat-line from last observed value), "continuedTrend" (model extrapolation), or "hybridTrend" (average of constant and continued). Default is "constantTrend".
group_cols: Character vector. Names of grouping columns (e.g., administrative units, insecticides). Default is c("adm0", "adm1").
value_col: Character scalar. Name of the mortality/resistance value column. Default is "mean_ir".

Value

A data frame with grouping columns, year, trendmethod, and the value column specified in value_col.

Examples

if (FALSE) { # \dontrun{
ir_data <- data.frame(
  adm0 = "BDI",
  adm1 = rep(c("Province A", "Province B"), each = 16),
  year = rep(2010:2025, 2),
  mean_ir = runif(32, 0.6, 0.95)
)

constant_proj <- run_resistance_trend(
  df_IR = ir_data,
  year_cut = 2025,
  n_years_to_add = 5,
  trendmethod = "constantTrend"
)

hybrid_proj <- run_resistance_trend(
  df_IR = ir_data,
  year_cut = 2025,
  trendmethod = "hybridTrend"
)
} # }