Bounded memory linear regression
bigglm.ffdf creates a generalized linear model object that uses only p^2 memory for p variables. It uses the biglm package and is a simple wrapper to allow to work with an ffdf as input data. Make sure that package is loaded.
bigglm.ffdf(formula, data, family = gaussian(), ..., chunksize = 5000)
formula |
a model formula |
data |
an object of class ffdf |
family |
A glm family object |
... |
other parameters passed on to bigglm. See the biglm package: |
chunksize |
Size of chunks for processing the ffdf |
An object of class bigglm. See the bigglm package for a description: bigglm
## Not run:
library(biglm)
library(ff)
data(trees)
x <- as.ffdf(trees)
a <- bigglm(log(Volume)~log(Girth)+log(Height),
data=x, chunksize=10, sandwich=TRUE)
summary(a)
b <- bigglm(log(Volume)~log(Girth)+log(Height)+offset(2*log(Girth)+log(Height)),
data=x, chunksize=10, sandwich=TRUE)
summary(b)
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.