Skip to content
/ hypr Public

hypr is an R package for easy translation between experimental (null) hypotheses and contrast matrices as used for linear regression.

License

Notifications You must be signed in to change notification settings

mmrabe/hypr

Repository files navigation

hypr

CRAN downloads JOSS publication status DOI Package status

hypr is an R package for easy translation between experimental (null) hypotheses and contrast matrices as used for linear regression. For an extensive overview of the package functions, see the hypr-intro vignette, e.g. by running vignette("hypr-intro") after installing the package.

Installation

Install from CRAN within R using:

install.packages("hypr")

Install the development version in R using devtools:

devtools::install_github("mmrabe/hypr", build_vignettes = TRUE)

Deriving contrast matrices

For a treatment contrast with a baseline and three treatment conditions:

trtC <- hypr(mu1~0, mu2~mu1, mu3~mu1, mu4~mu1)
trtC
## hypr object containing 4 null hypotheses:
## H0.1: 0 = mu1        (Intercept)
## H0.2: 0 = mu2 - mu1
## H0.3: 0 = mu3 - mu1
## H0.4: 0 = mu4 - mu1
## 
## Call:
## hypr(~mu1, ~mu2 - mu1, ~mu3 - mu1, ~mu4 - mu1, levels = c("mu1", 
## "mu2", "mu3", "mu4"))
## 
## Hypothesis matrix (transposed):
##     [,1] [,2] [,3] [,4]
## mu1  1   -1   -1   -1  
## mu2  0    1    0    0  
## mu3  0    0    1    0  
## mu4  0    0    0    1  
## 
## Contrast matrix:
##     [,1] [,2] [,3] [,4]
## mu1 1    0    0    0   
## mu2 1    1    0    0   
## mu3 1    0    1    0   
## mu4 1    0    0    1

To assign the contrast matrix to a factor fac with an intermediate hypr object:

contrasts(fac) <- contr.hypothesis(trtC)

… or without an intermediate object:

contrasts(fac) <- contr.hypothesis(mu1~0, mu2~mu1, mu3~mu1, mu4~mu1) 

For more information, see vignette("hypr-regression").

Deriving research hypotheses

To check which reserach (null) hypotheses a given contrast matrix is testing, we can create an empty hypr object.

testC <- hypr() # create an empty hypr object

A treatment contrast with 4 levels (incl. the baseline) can look as follows:

contr.treatment(4)
##   2 3 4
## 1 0 0 0
## 2 1 0 0
## 3 0 1 0
## 4 0 0 1

We can now populate the hypr object by setting its contrast matrix. Note that the treatment contrast does not have an intercept. We thus have to add it when populating the hypr object:

cmat(testC, add_intercept = TRUE) <- contr.treatment(4) # populate object via contrast matrix

Now, the hypr object contains 4 hypotheses, a hypothesis matrix and the contrast matrix identical to the treatment contrast with an intercept added:

testC
## hypr object containing 4 null hypotheses:
## H0.Intercept: 0 = X1        (Intercept)
##         H0.2: 0 = -X1 + X2
##         H0.3: 0 = -X1 + X3
##         H0.4: 0 = -X1 + X4
## 
## Call:
## hypr(Intercept = ~X1, `2` = ~-X1 + X2, `3` = ~-X1 + X3, `4` = ~-X1 + 
##     X4, levels = c("X1", "X2", "X3", "X4"))
## 
## Hypothesis matrix (transposed):
##    Intercept 2  3  4 
## X1  1        -1 -1 -1
## X2  0         1  0  0
## X3  0         0  1  0
## X4  0         0  0  1
## 
## Contrast matrix:
##    Intercept 2 3 4
## X1 1         0 0 0
## X2 1         1 0 0
## X3 1         0 1 0
## X4 1         0 0 1

The derived hypotheses can be rewritten as:

For more information, see vignette("hypr-contrasts").

Community guidelines

If you want to report a bug, are having technical difficulties or want to recommend features, it’s best to open a Github Issue. If you want to suggest a specific implementation of a feature or bug fix, you’re welcome to fork the repository and submit a pull request! Alternatively, if you are having problems or questions, you can also send an e-mail (maximilian.rabe@uni-potsdam.de).

About

hypr is an R package for easy translation between experimental (null) hypotheses and contrast matrices as used for linear regression.

Resources

License

Stars

Watchers

Forks

Packages

No packages published