scidb (version 1.1-2)

predict.glm_scidb: Prediction for SciDB GLM fits

Description

Obtains predictions and optionally estimates standard errors of those predictions from a fitted generalized linear model object.

Usage

## S3 method for class 'glm_scidb':
predict(object, ...)

Arguments

object

a glm_scidb model object.

...

optional arguments newdata, type, and se.fit (see details).

Value

• If se.fit = FALSE, a vector of predictions.

  If se.fit = TRUE, a list with components

• fit
  Predictions, as for se.fit = FALSE.
• se.fit
  Estimated standard errors.
• residual.scale
  A scalar giving the square root of the dispersion used in computing the standard errors.

Details

Optional arguments:

• newdata

{ a scidbdf SciDB data frame in which to look for variables with which to predict. If omitted, the fitted linear predictors are used.} type{ the type of prediction required. The default is on the scale of the linear predictors; the alternative "response" is on the scale of the response variable. Thus for a default binomial model the default predictions are of log-odds (probabilities on logit scale) and type = "response" gives the predicted probabilities.} se.fit{ logical switch indicating if standard errors are required.}

See Also

scidb glm_scidb

Examples

## example adpted from Venables and Ripley (2002, pp. 190-2.):

# In R:
ldose   <- rep(0:5, 2)
numdead <- c(1, 4, 9, 13, 18, 20, 0, 2, 6, 10, 12, 16)
sex     <- factor(rep(c("M", "F"), c(6, 6)))
data    <- data.frame(sex, ldose)
data    <- Reduce(rbind, 
             lapply(1:length(numdead),
               function(j) rbind(cbind(alive=1,data[j,])[rep(1,numdead[j]),],
                                 cbind(alive=0,data[j,])[rep(1,20-numdead[j]),])))
rownames(data) <- NULL

r_model <- glm( alive ~ sex + ldose - 1, family=binomial(), data=data)

# Now in SciDB:
data_scidb <- as.scidb(data)
str(data_scidb)
scidb_model <- glm_scidb( alive ~ sex + ldose - 1, family=binomial(), data=data_scidb)

# New data for prediction:
ld <- seq(0,5,0.1)
newdata <- as.scidb(data.frame(ldose=ld, sex=rep("M",length(ld))))
head(newdata)

pred_scidb = predict(scidb_model, newdata=newdata, type="response")
head(pred_scidb)

require("graphics")
plot(c(1,32), c(0,1), type = "n", xlab = "dose",
          ylab = "prob", log = "x")
text(2^ldose, numdead/20, as.character(sex))
lines(2^ld, pred_scidb[],lwd=2,col=4)