coefeasy

Coefeasy is an R package under development for making regression coefficients more accessible. With this tool, you can read and report key coefficients instantly.

https://github.com/mateoservent/coefeasy

Keywords

Repository

Coefeasy is an R package under development for making regression coefficients more accessible. With this tool, you can read and report key coefficients instantly.

README.Rmd

---
output: github_document
---



# Coefeasy 


![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)
[![CRAN status](https://www.r-pkg.org/badges/version/Coeffeasy)](https://CRAN.R-project.org/package=Coeffeasy)


`Coefeasy` is a simple R package under development for making regression coefficients more accessible. With this tool, you can read linear and logistic regression coefficients accurately and instantly. The output of the functions here is only readings in natural language and helps to standardize the phrasing of widely used models. The package offers results not only in English, but in multiple languages, and can automatically control for heteroscedasticity and offer the capability to choose the type of standard error for the correction and group errors by cluster—all in a single function.

`Coefeasy` is a straightforward package designed primarily for, but not limited to, people learning or teaching introduction to statistics or econometrics. Historically, although an interpretation of coefficients should follow a consistent design, the underlying knowledge to read the results of a regression can be quite technical and unambiguous in its presentation. Acknowledging this challenge, `coefeasy` offers a systematic solution, primarily serving linear and logistic models. As evidence-based decision making continues to gain importance, `coefeasy` aims to help close that knowledge gap, providing clear and accessible statistical learning, and reinforcing the importance of automation and standardization.


## Installation

Coeffeasy can be installed using the development version of the package from [github](https://github.com/your_github_username/Coeffeasy) with:


``` r
install.packages("remotes")
remotes::install_github("mateoservent/coefeasy")

```
 
## Example Linear Regression

In this example, we use `coefeasy_lm()` to interpret coefficients from a linear regression on the mtcars dataset, analyzing how car weight affects miles per gallon when controlling for horsepower and quarter mile time.

``` r
library(coefeasy)

# Fit the linear regression model
model <- lm(mpg ~ wt + hp + qsec, data = mtcars)

# Use Coeffeasy's function to interpret the coefficients

# Using the default settings, which automatically deduce variable names and use an alpha of 0.05
coefeasy_lm(model) 

# Specifying the names of the response and predictor variables for a clearer interpretation, adjusting the significance level, and selecting a longer, more descriptive result
coefeasy_lm(model, y = "Miles per gallon", x = "Car weight (1000 lbs)", alpha = 0.01, short = F) 

# To display results in another language, such as French, use the argument 'language="fr"' (beta).

```
## Example Logistic Regression

Second example, we use `coefeasy_logit()` to interpret coefficients from a logistic regression on the same dataset. We will analyze how the weight of the car affects the likelihood of a car having an automatic transmission (1 = Automatic; 0 = Manual).

``` r
# Fit the logistic regression model
model_logit <- glm(am ~ wt, data = mtcars, family = "binomial")

# Using the default settings, which automatically deduce variable names and use an alpha of 0.05
coefeasy_logit(model_logit) 

# Specifying the names of the response and predictor variables for a clearer interpretation, and adjusting the significance level
coefeasy_logit(model_logit, y = "Automatic Transmission", x = "Car weight (1000 lbs)", alpha = 0.01) 
```

Owner

CodeMeta (codemeta.json)

{
  "@context": "https://doi.org/10.5063/schema/codemeta-2.0",
  "@type": "SoftwareSourceCode",
  "identifier": "Coefeasy",
  "description": "The package offers tools to interpret statistical model results in natural language. It covers linear and logit models, evaluates model assumptions such as heteroscedasticity, and presents results in a comprehensible manner. Users can choose between different languages for output, enhancing the accessibility and understanding of complex analyses.",
  "name": "Coefeasy: Coefeasy is an R package for making regression coefficients more accessible. With this tool, you can plot and read many key coefficients instantly",
  "license": "https://spdx.org/licenses/MIT",
  "version": "0.0.0.9000",
  "programmingLanguage": {
    "@type": "ComputerLanguage",
    "name": "R",
    "url": "https://r-project.org"
  },
  "runtimePlatform": "R version 4.3.3 (2024-02-29)",
  "author": [
    {
      "@type": "Person",
      "givenName": "Mateo",
      "familyName": "Servent",
      "email": "mateoservent@gmail.com"
    }
  ],
  "maintainer": [
    {
      "@type": "Person",
      "givenName": "Mateo",
      "familyName": "Servent",
      "email": "mateoservent@gmail.com"
    }
  ],
  "softwareRequirements": {
    "1": {
      "@type": "SoftwareApplication",
      "identifier": "lmtest",
      "name": "lmtest",
      "provider": {
        "@id": "https://cran.r-project.org",
        "@type": "Organization",
        "name": "Comprehensive R Archive Network (CRAN)",
        "url": "https://cran.r-project.org"
      },
      "sameAs": "https://CRAN.R-project.org/package=lmtest"
    },
    "2": {
      "@type": "SoftwareApplication",
      "identifier": "sandwich",
      "name": "sandwich",
      "provider": {
        "@id": "https://cran.r-project.org",
        "@type": "Organization",
        "name": "Comprehensive R Archive Network (CRAN)",
        "url": "https://cran.r-project.org"
      },
      "sameAs": "https://CRAN.R-project.org/package=sandwich"
    },
    "3": {
      "@type": "SoftwareApplication",
      "identifier": "polyglotr",
      "name": "polyglotr",
      "provider": {
        "@id": "https://cran.r-project.org",
        "@type": "Organization",
        "name": "Comprehensive R Archive Network (CRAN)",
        "url": "https://cran.r-project.org"
      },
      "sameAs": "https://CRAN.R-project.org/package=polyglotr"
    },
    "SystemRequirements": null
  },
  "fileSize": "140.04KB"
}

GitHub Events