ICEinfer - Incremental Cost-Effectiveness Inference using Two Unbiased Samples

Given two unbiased samples of patient level data on cost and effectiveness for a pair of treatments, make head-to-head treatment comparisons by (i) generating the bivariate bootstrap resampling distribution of ICE uncertainty for a specified value of the shadow price of health, lambda, (ii) form the wedge-shaped ICE confidence region with specified confidence fraction within [0.50, 0.99] that is equivariant with respect to changes in lambda, (iii) color the bootstrap outcomes within the above confidence wedge with economic preferences from an ICE map with specified values of lambda, beta and gamma parameters, (iv) display VAGR and ALICE acceptability curves, and (v) illustrate variation in ICE preferences by displaying potentially non-linear indifference(iso-preference) curves from an ICE map with specified values of lambda, beta and either gamma or eta parameters.

Last updated 4 years ago

1.76 score 19 scripts 334 downloads

RXshrink - Maximum Likelihood Shrinkage using Generalized Ridge or Least Angle Regression

Functions are provided to calculate and display ridge TRACE Diagnostics for a variety of alternative Shrinkage Paths. While all methods focus on Maximum Likelihood estimation of unknown true effects under normal distribution-theory, some estimates are modified to be Unbiased or to have "Correct Range" when estimating either [1] the noncentrality of the F-ratio for testing that true Beta coefficients are Zeros or [2] the "relative" MSE Risk (i.e. MSE divided by true sigma-square, where the "relative" variance of OLS is known.) The eff.ridge() function implements the "Efficient Shrinkage Path" introduced in Obenchain (2022) <Open Statistics>. This "p-Parameter" Shrinkage-Path always passes through the vector of regression coefficient estimates Most-Likely to achieve the overall Optimal Variance-Bias Trade-Off and is the shortest Path with this property. Functions eff.aug() and eff.biv() augment the calculations made by eff.ridge() to provide plots of the bivariate confidence ellipses corresponding to any of the p*(p-1) possible ordered pairs of shrunken regression coefficients. Functions for plotting TRACE Diagnostics now have more options.

Last updated 2 years ago

1.18 score 15 scripts 517 downloads

NU.Learning - Nonparametric and Unsupervised Learning from Cross-Sectional Observational Data

Especially when cross-sectional data are observational, effects of treatment selection bias and confounding are best revealed by using Nonparametric and Unsupervised methods to "Design" the analysis of the given data ...rather than the collection of "designed data". Specifically, the "effect-size distribution" that best quantifies a potentially causal relationship between a numeric y-Outcome variable and either a binary t-Treatment or continuous e-Exposure variable needs to consist of BLOCKS of relatively well-matched experimental units (e.g. patients) that have the most similar X-confounder characteristics. Since our NU Learning approach will form BLOCKS by "clustering" experimental units in confounder X-space, the implicit statistical model for learning is One-Way ANOVA. Within Block measures of effect-size are then either [a] LOCAL Treatment Differences (LTDs) between Within-Cluster y-Outcome Means ("new" minus "control") when treatment choice is Binary or else [b] LOCAL Rank Correlations (LRCs) when the e-Exposure variable is numeric with (hopefully many) more than two levels. An Instrumental Variable (IV) method is also provided so that Local Average y-Outcomes (LAOs) within BLOCKS may also contribute information for effect-size inferences when X-Covariates are assumed to influence Treatment choice or Exposure level but otherwise have no direct effects on y-Outcomes. Finally, a "Most-Like-Me" function provides histograms of effect-size distributions to aid Doctor-Patient (or Researcher-Society) communications about Heterogeneous Outcomes. Obenchain and Young (2013) <doi:10.1080/15598608.2013.772821>; Obenchain, Young and Krstic (2019) <doi:10.1016/j.yrtph.2019.104418>.

Last updated 1 years ago

1.00 score 2 scripts 172 downloads