Learn R Programming

sisireg (version 1.2.1)

Sign-Simplicity-Regression-Solver

Description

Implementation of the SSR-Algorithm. The Sign-Simplicity-Regression model is a nonparametric statistical model which is based on residual signs and simplicity assumptions on the regression function. Goal is to calculate the most parsimonious regression function satisfying the statistical adequacy requirements. Theory and functions are specified in Metzner (2020, ISBN: 979-8-68239-420-3, "Trendbasierte Prognostik") and Metzner (2021, ISBN: 979-8-59347-027-0, "Adäquates Maschinelles Lernen").

Copy Link

Version

Install

install.packages('sisireg')

Monthly Downloads

214

Version

1.2.1

License

GPL (>= 2)

Maintainer

Lars Metzner

Last Published

July 9th, 2025

Functions in sisireg (1.2.1)

Multi-dimensional SSR model - reworked

Near Neighborhood Estimator for Multi-dimensional SSR model

Partial Sum Plot

Onedimensional SSR-model calculation

Prediction function for the multi-dimensional SSR model - reworked

2-layer MLP with partial sum optimization

3-dimensional SSR model prediction

SSR model Prediction

Multi-dimensional SSR model

Trend-based Correlation

3-dimensional SSR model

Prediction function for the multi-dimensional SSR model

Factor-wise Influence Indicator (Prediction-fii) for a given ssrmlp model regarding a given input vector

axe_narch_model

implementation of the AxE model based on the ssr-MLP

Saving a ssrmlp model in onnx format to file

Partial Sum Validity Check

Partial Sum Plot for 2-dimensional coordinates

Loading a ssrmlp model from ONNX file

axe_narch_predict

Prediction function for the AxE-NARCH model

Partial Sum Plot for the multidimensional coordinates

Factor-wise Influence Indicator (Model-fii) for a given ssrmlp model

Data model for the AxE-Model (Axiomatic Econometric Modeling Paradigm)

Maximum Run Validity Check

Partial Sum Plot for the multidimensional coordinates - reworked

2-layer MLP with partial sum optimization - reworked

Prediction function for the ssrMLP