Search Results for "estimate"

Bayes+Estimate is a Rust and C++ library that implement numerical algorithms for Bayesian estimation. They provide tested and consistent numerical methods and represents the wide variety of Bayesian estimation algorithms and system model.

The CausalImpact repository houses an R package that implements causal inference in time series using Bayesian structural time series models. Its goal is to estimate the effect of an intervention (e.g. a marketing campaign, policy change) on a time series outcome by predicting what would have happened in a counterfactual “no intervention” world. The package requires as input a response time series plus one or more control (covariate) time series that are assumed unaffected by the intervention, and it divides the time horizon into “pre-intervention” and “post-intervention” periods. ...

...The first version of this cybernetic approach had been developed for the evaluation of the functional status of the thyroid gland, i.e. thyroid's secretory capacity (gain of thyroid, GT or SPINA-GT) and sum activity of peripheral deiodinases (GD or SPINA-GD). A second version has been established for insulin-glucose homeostasis. It delivers the secretory capacity of pancreatic beta cells (SPINA-GBeta), an estimate for the insulin receptor amplification (SPINA-GR) and a static disposition index as a measure for the loop gain of the feedback loop (SPINA-DI). Both implementations of SPINA have been validated in more than 50 clinical studies based on large cohorts.

Geometric Regression Library (GeoRegression) is an open source Java geometry library for scientific computing with a focus on 2D/3D space. GeoRegression provides the ability to estimate the closest point/distance between geometric primitives, best-fit shapes, and best fit geometric transform between sets of objects. It is designed for high performance and ease of use. GeoRegression has been release under an LGPL license for both commercial and non-commercial use.

The new site is at http://psychometricon.net/libirt/ Library of functions to estimate the items and abilities from the responses of subjects to a questionnaire. The IRT models supported are the logistic model, the multivariate logistic model, the graded model and smoothing by penalization and kernel.

An EDM calibration baseline is used to derive specific instrument errors of the electronic distance measurement (EDM) unit of a total station. EDMFit is a simple adjustment tool to estimate the EDM calibration parameters (addition constant and/or scale parameter). The adjustment process is according to ISO 17123-4. The parameters are checked for significance, to avoid an overestimation. Based on two classical hypothesis testing, EDMFit provides a reliable outlier detection. For uncertainties budgeting, EDMFit makes use of Monte-Carlo simulations and supports different distribution function: - Uniform distribution, - Triangular distribution and - Gaussian (normal) distribution. ...

...Although JBernstein uses double, it offers automatic error-estimation capabilities (experimental) so that the generated result is correct (the solver may return unknown when the error-estimate is too large)

...Mixed models partition mortality into exogenous and endogenous components, so that the intrinsic survivorship can be estimated without the interference from extrinsic noise. DeDAY supports both interval-censored data and exact event-time data. Using MLE (Maximum Likelihood Estimate), DeDAY fits statistic model to the data. DeDAY also calculates the variances and the multi-dimensional confidence limits of model parameters. DeDAY is free for academic users.

The Felo rating system provides an objective fencing strength estimate for each individual fencer in a group, just like the Elo rating in chess. This program calculates the Felo numbers by analysing a bout result list created by the user.

A MATLAB package to simulate sample paths of the solution of a Itô or Stratonovich stochastic differential equation (SDE), compute statistics and estimate the parameters from data. A note of caution: SDE Toolbox is no more developed but it's still downloadable. Its inferential capabilities can be considered surpassed (at best). Actually the parameter estimation methods were already far from the state-of-art when the project began in 2007 (!). The considered implemented parametric and non-parametric Monte Carlo likelihood methods were chosen for their ability to treat both one-dimensional and multivariate SDE systems, although the quality of the inferential results can't match those obtained using more advanced techniques. ...

Algorithms to estimate the number of k-colorings of a low-degree graph using samplings from a markov chain are provided.

Numerically estimate an efficiency of code construction algorithms. Randomly generate n binary words of length m with minimum Hamming distance d and create a prefix code (= retrieval tree) from them. Based on PhD work.