bestglm
(Package: bestglm) :
Best Subset GLM using Information Criterion or Cross-Validation
Best subset selection using 'leaps' algorithm (Furnival and Wilson, 1974) or complete enumeration (Morgan and Tatar, 1972). Complete enumeration is used for the non-Gaussian and for the case where the input matrix contains factor variables with more than 2 levels. The best fit may be found using the information criterion IC: AIC, BIC, EBIC, or BICq. Alternatively, with IC=‘CV’ various types of cross-validation may be used.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: bestglm
●
0 images
|
print.bestglm
(Package: bestglm) :
Print method for `bestglm` object
A brief description of the best fit is given.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: print.bestglm
●
0 images
|
asbinary
(Package: bestglm) :
Binary representation of non-negative integer
A non-negative integer is represented as a binary number. The digits, 0 or 1, of this number are returned in a vector.
● Data Source:
CranContrib
● Keywords: arith
● Alias: to.binary
●
0 images
|
CVHTF
(Package: bestglm) :
K-fold Cross-Validation
K-fold cross-validation.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: CVHTF
●
0 images
|
|
An analysis of deviance and a likelihood-ratio test with p-value. The p-value is greatly exagerated due to selection.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: summary.bestglm
●
0 images
|
oneSdRule
(Package: bestglm) :
Utility function. Implements the 1-sd rule.
The CV and its standard devation are provided for a range of models ordered by the number of parameters estimated.
● Data Source:
CranContrib
● Keywords:
● Alias: oneSdRule
●
0 images
|
CVd
(Package: bestglm) :
Cross-validation using delete-d method.
The delete-d method for cross-validation uses a random sample of d observations as the validation sample. This is repeated many times.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: CVd
●
0 images
|
|
Provides new information criterion BICq as well as AIC, BIC and EBIC for selecting the best model. Additionally, various CV algorithms are also provided.
● Data Source:
CranContrib
● Keywords: package
● Alias: bestglm-package
●
0 images
|
LOOCV
(Package: bestglm) :
Leave-one-out cross-validation
An observation is removed and the model is fit the the remaining data and this fit used to predict the value of the deleted observation. This is repeated, n times, for each of the n observations and the mean square error is computed.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: LOOCV
●
0 images
|
CVDH
(Package: bestglm) :
Adjusted K-fold Cross-Validation
An adjustment to K-fold cross-validation is made to reduce bias.
● Data Source:
CranContrib
● Keywords: models, regression
● Alias: CVDH
●
0 images
|
providing 
.