REG: Regression of Linear Least Square, similar to SAS PROC REG

Description

REG is similar to SAS PROC REG.

Usage

REG(Formula, Data, conf.level=0.95, HC=FALSE, Weights=1, summarize=TRUE)

Value

The result is comparable to that of SAS PROC REG.

The first part is ANOVA table.

The second part is measures about fitness.

The third part is the esimates of coefficients.

Estimate: point estimate of parameters, coefficients
Estimable: estimability: 1=TRUE, 0=FALSE. This appears only when at least one inestimability occurs.
Std. Error: standard error of the point estimate
Lower CL: lower confidence limit with conf.level
Upper CL: lower confidence limit with conf.level
Df: degree of freedom
t value: value for t distribution
Pr(>|t|): probability of larger than absolute t value from t distribution with residual's degree of freedom

The above result is repeated using HC0 and HC3, with following White's first and second moment specification test, if HC option is specified. The statstic t values and their p values with HC1 and HC2 are between those of HC0 and H3.

If summarize=FALSE, REG returns;

coeffcients: beta coefficients
g2: g2 inverse
rank: rank of the model matrix
DFr: degree of freedom for the residual
SSE: sum of square error

Arguments

Formula: a conventional formula for a linear model
Data: a data.frame to be analyzed
conf.level: confidence level for the confidence limit
HC: heteroscedasticity related output is required such as HC0, HC3, White's first and second moment specification test
Weights: weights for the each observation or residual square. This is usually the inverse of each variance.
summarize: If this is FALSE, REG returns just lfit result.

Author

Kyun-Seop Bae k@acr.kr

Details

It performs the core function of SAS PROC REG.

Examples

Run this code

  REG(uptake ~ Plant + Type + Treatment + conc, CO2)
  REG(uptake ~ conc, CO2, summarize=FALSE)

Run the code above in your browser using DataLab