Skip to contents

LAGO 1.0.12

  • lago_optimization() now returns an object of class "lago" with print(), summary(), and plot() methods for a readable console summary and a confidence-set plot. The result is still a plain list, so existing $-based access is unchanged. (#61)
  • Added an “Optimizing an intervention with LAGO” vignette that walks through a full optimization on the bundled BetterBirth data, rendered as an Articles page on the documentation site. (#61)
  • Added a hex logo and refreshed the README header, badges plus a title logo, and removed the now-redundant banner. (#56, #61)
  • Added a quiet argument to lago_optimization() that skips the progress messages and their paced delays. Programmatic and repeated calls run substantially faster while returning identical results. (#60)
  • visualize_cost() gained a “Copy to clipboard” button, a numeric cost summary, and aligned cost and marginal-cost axes. On close it now returns the assembled cost list to the R session, assigned to lago_cost_list. (#59)
  • Expanded the reference documentation for the bundled datasets with fuller variable descriptions. (#58)
  • Added a pkgdown documentation site published to GitHub Pages. (#53)
  • Added a GPL-3 LICENSE file and polished the README presentation. (#56)
  • Fixed a crash when a single intervention component was used. (#54)
  • Added an icc argument to the power calculation that applies a design-effect adjustment, so the power goal can account for within-center clustering. (#29)
  • The overall intervention-effect test result (statistic and p-value) is now returned in the result object rather than only printed. (#49)
  • Added an automated test suite (testthat) and continuous-integration checks. (#45)
  • visualize_cost() sliders now support per-component default and custom ranges. (#34)
  • Added non-fatal outcome-model fit diagnostics that warn about a questionable fit while the optimization continues. (#36)
  • Supplying optimization_grid_search_step_size now switches the optimization method to grid search automatically. (#32)
  • lago_optimization() now accepts a standalone power goal, an outcome goal, or both together, taking the higher of the outcome goal and the power-implied outcome. (#40)