ag.ds.glm: Runs a combined GLM analysis of non-pooled data

Description Usage Arguments Details Value Author(s) Examples

Description

Runs a combined GLM analysis of non-pooled data

Usage

1
  ag.ds.glm(opals, formula, family, maxit = 10)

Arguments

opals

a list of opal object(s) obtained after login in to opal servers; these objects hold also the data assign to R, as dataframe, from opal datasources.

formula

an object of class formula which describes the model to be fitted

family

a description of the error distribution function to use in the model

maxit

the number of iterations of IWLS used

Details

It enables a parallelized analysis of individual-level data sitting on distinct servers by sending instructions to each computer requesting non-disclosing summary statistics. The sumaries are then combined to estimate the parameters of the model; these parameters are the same as those obtained if the data were 'physically' pooled.

Value

coefficients a named vector of coefficients

residuals the 'working' residuals, that is the residuals in the final iteration of the IWLS fit.

fitted.values the fitted mean values, obtained by transforming the linear predictors by the inverse of the link function.

rank the numeric rank of the fitted linear model.

family the family object used.

linear.predictors the linear fit on link scale.

aic A version of Akaike's An Information Criterion, which tells how well the model fits

Author(s)

Burton, P.; Laflamme, P.; Gaye, A.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
{
# load the file that contains the login details
data(logindata)

# login and assign some variables to R
myvar <- list("DIS_DIAB","PM_BMI_CONTINUOUS","LAB_HDL")
opals <- ag.ds.login(logins=logindata,assign=TRUE,variables=myvar)

# run a GLM (e.g. diabetes prediction using BMI and HDL level)
 mod <- ag.ds.glm(opals=opals,formula=D$DIS_DIAB~D$PM_BMI_CONTINUOUS+D$LAB_HDL,family=quote(binomial))
}

datashield/ag.dev.cl documentation built on May 14, 2019, 7:46 p.m.