relevance (version 2.1)

inference: Calculate Confidence Intervals and Relevance and Significance Values

Description

Calculates confidence intervals and relevance and significance values given estimates, standard errors and, for relevance, additional quantities.

Usage

inference(object = NULL, estimate = NULL, teststatistic = NULL,
  se = NA, n = NULL, df = NULL,
  stcoef = TRUE, rlv = TRUE, rlv.threshold = getOption("rlv.threshold"),
  testlevel = getOption("testlevel"), ...)

Value

A data.frame of class "inference", with the variables

effect, se

estimated effect(s), often coefficients, and their standard errors

ciLow, ciUp

lower and upper limit of the confidence interval

teststatistic

t-test statistic

p.value

p value

Sig0

significance value, i.e., test statistic divided by critical value, which in turn is the 1-testlevel/2-quantile of the t-distribution.

ciLow, ciUp

confidence interval for effect

If rlv is TRUE,

stcoef

standardized coefficient

st.Low, st.Up

confidence interval for stcoef

Rle

estimated relevance of coef

Rls

secured relevance, lower end of confidence interval for the relevance of coef

Rlp

potential relevance, upper end of confidence interval ...

Rls.symbol

symbols for the secured relevance

Rlvclass

relevance class

Arguments

object

A data.frame containing, as its variables, the arguments estimate to df, as far as needed, or a vector to be used as estimate if estimate is not specified...

... or a model fit object

estimate

estimate(s) of the parameter(s)

teststatistic

test statistic(s)

se

standard error(s) of the estimate(s)

n

number(s) of observations

df

degrees of freedom of the residuals

stcoef

standardized coefficients. If NULL, these will be calculated from object, if the latter is a model fit.

rlv

logical: Should relevances be calculated?

rlv.threshold

Relevance threshold(s). May be a simple number for simple inference, or a vector containing the elements

stand:

threshold for (simple) standardized effects

rel:

for relative effects,

coef:

for standardized coefficients,

drop:

for drop effects,

pred:

for prediction intervals.

testlevel

1 - confidence level

...

furter arguments, passed to termtable and termeffects

Author

Werner A. Stahel

Details

The estimates divided by standard errors are assumed to be t-distributed with df degrees of freedom. For df==Inf, this is the standard normal distribution.

References

Werner A. Stahel (2020). New relevance and significance measures to replace p-values. PLOS ONE 16, e0252991, doi: 10.1371/journal.pone.0252991

See Also

link{twosamples}, link{termtable}, link{termeffects}

Examples

data(d.blast)
rr <-
  lm(log10(tremor)~location+log10(distance)+log10(charge),
    data=d.blast) 
inference(rr)