Help for package BetterReg

Type:

Package

Title:

Better Statistics for OLS and Binomial Logistic Regression

Version:

0.2.0

Author:

Chris Aberson

Maintainer:

Chris Aberson <cla18@humboldt.edu>

Description:

Provides squared semi partial correlations, tolerance, Mahalanobis, Likelihood Ratio Chi Square, and Pseudo R Square. Aberson, C. L. (2022) <doi:10.31234/osf.io/s2yqn>.

License:

GNU General Public License version 3

Encoding:

UTF-8

LazyData:

true

Imports:

car (≥ 3.0-0), stats (≥ 3.5.0), dplyr (≥ 0.8.0)

Depends:

R (≥ 3.5.0)

RoxygenNote:

7.1.2

NeedsCompilation:

Packaged:

2022-03-29 23:41:52 UTC; cla18

Repository:

CRAN

Date/Publication:

2022-03-30 17:00:01 UTC

Compute Likelihood Ratio Chi-square for Binomial Logistic Regression with up to 10 predictors

Description

Compute Likelihood Ratio Chi-square for Binomial Logistic Regression with up to 10 predictors

Usage

LRchi(
  data = NULL,
  y = NULL,
  x1 = NULL,
  x2 = NULL,
  x3 = NULL,
  x4 = NULL,
  x5 = NULL,
  x6 = NULL,
  x7 = NULL,
  x8 = NULL,
  x9 = NULL,
  x10 = NULL,
  numpred = NULL
)

Arguments

data

name of your datafile, loaded

y

dependent variable name

x1

first predictor variable name

x2

second predictor variable name

x3

third predictor variable name

x4

fourth predictor variable name

x5

fifth predictor variable name

x6

sixth predictor variable name

x7

seventh predictor variable name

x8

eighth predictor variable name

x9

ninth predictor variable name

x10

tenth predictor variable name

numpred

number of predictors

Examples

LRchi(data=testlog, y="dv", x1="iv1", x2="iv2",numpred=2)

Compute Mahalanobis Distance for Multiple Regression

Description

Compute Mahalanobis Distance for Multiple Regression

Usage

Mahal(model = NULL, pred = NULL, values = 5)

Arguments

model

name of model

pred

number of predictors

values

number of Mahal values to print (highest values). Default is 10

Value

Mahalanobis Distance to detect MV outliers

Examples

mymodel<-lm(y~x1+x2+x3+x4, testreg)
Mahal(model=mymodel, pred=5, values = 10)

R-square change for Hierarchical Multiple Regression

Description

R-square change for Hierarchical Multiple Regression

Usage

R2change(model1 = NULL, model2 = NULL)

Arguments

model1

first regression model

model2

second regression model

Examples

mymodel1<-lm(y~x1+x2, data=testreg)
mymodel2<-lm(y~x1+x2+x3+x4, data=testreg)
R2change(model1=mymodel1, model2=mymodel2)

Power for Comparing Dependent Coefficients in Multiple Regression with Two or Three Predictors Requires correlations between all variables as sample size. Means, sds, and alpha are option. Also computes Power(All)

Description

Power for Comparing Dependent Coefficients in Multiple Regression with Two or Three Predictors Requires correlations between all variables as sample size. Means, sds, and alpha are option. Also computes Power(All)

Usage

depbcomp(
  data = NULL,
  y = NULL,
  x1 = NULL,
  x2 = NULL,
  x3 = NULL,
  x4 = NULL,
  x5 = NULL,
  numpred = NULL,
  comps = "abs"
)

Arguments

data

name of data file

y

dependent variable name

x1

first predictor variable name

x2

second predictor variable name

x3

third predictor variable name

x4

fourth predictor variable name

x5

fifth predictor variable name

numpred

number of predictors

comps

Type of comparison, "abs" for absolute values or "raw" for raw coefficients

Value

Comparing Dependent Coefficients in Multiple Regression

Examples

depbcomp(data=testreg,y=y,x1=x1,x2=x2,x3=x3,x4=x4,x5=x5, numpred=5,comps="abs")

Comparing Independent Coefficients in Multiple Regression

Description

Comparing Independent Coefficients in Multiple Regression

Usage

indbcomp(model1 = NULL, model2 = NULL, comps = "abs")

Arguments

model1

Summary of first model (see example for how to summarize)

model2

Summary of second model (see example for how to summarize)

comps

Type of comparison. "abs" - absolute value of coefficient (recommended). "raw" raw values of coefficient

Value

Comparing Independent Coefficients in Multiple Regression

Examples

y_1<-rnorm(200); x1_1<-rnorm(200); x2_1<-rnorm(200)
y_2<-rnorm(200); x1_2<-rnorm(200);x2_2<-rnorm(200)
df1<-as.data.frame(cbind(y_1, x1_1,x2_1))
df2<-as.data.frame(cbind(y_2, x1_2,x2_2))
model1_2<-summary(lm(y_1~x1_1+x2_1, data=df1))
model2_2<-summary(lm(y_2~x1_2+x2_2, data=df2))
indbcomp(model1 = model1_2, model2 = model2_2, comps="abs")

Compute squared semi partial correlations for Multiple Regression

Description

Compute squared semi partial correlations for Multiple Regression

Usage

parts(model = NULL, pred = NULL)

Arguments

model

name of model

pred

number of predictors

Value

Squared semipartial correlations for MRC with up to 10 predictors

Examples

mymodel<-lm(y~x1+x2+x3+x4+x5, data=testreg)
parts(model=mymodel, pred=5)

Pseudo R-square Values for Binomial Logistic Regression

Description

Pseudo R-square Values for Binomial Logistic Regression

Usage

pseudo(model = NULL)

Arguments

model

name of model

Value

Pseudo R-square Values for Logistic Regression

Examples

mymodel<-glm(dv~iv1+iv2+iv3+iv4, testlog,family = binomial())
pseudo(model=mymodel)

testlog

Description

A dataset to test logistic regression functions

Usage

testlog

Format

A data frame with 164 rows and 11 variables:

dv: DV
iv1: 1st predictor
iv2: 2nd predictor
iv3: 3rd predictor
iv4: 4th predictor
iv5: 5th predictor
iv6: 6th predictor
iv7: 7th predictor
iv8: 8th predictor
iv9: 9th predictor
iv10: 10th predictor

testreg

Description

A dataset to test regression functions

Usage

testreg

Format

A data frame with 1000 rows and 6 variables:

y: DV
x1: 1st predictor
x2: 2nd predictor
x3: 3rd predictor
x4: 4th predictor
x5: 5th predictor

Compute tolerance for Multiple Regression

Description

Compute tolerance for Multiple Regression

Usage

tolerance(model = NULL)

Arguments

model

name of model

Value

Tolerance for MR

Examples

mymodel<-lm(y~x1+x2+x3+x4+x5, data=testreg)
tolerance(model=mymodel)