gss: Interfaces for gss package for data science pipelines.

In intubate: Interface to Popular R Functions for Data Science Pipelines

Description

Interfaces to gss functions that can be used in a pipeline implemented by magrittr.

Usage

ntbt_gssanova(data, ...)
ntbt_gssanova0(data, ...)
ntbt_gssanova1(data, ...)
ntbt_ssanova(data, ...)
ntbt_ssanova0(data, ...)
ntbt_ssanova9(data, ...)
ntbt_sscden(data, ...)
ntbt_sscden1(data, ...)
ntbt_sscox(data, ...)
ntbt_ssden(data, ...)
ntbt_ssden1(data, ...)
ntbt_sshzd(data, ...)
ntbt_ssllrm(data, ...)

Arguments

data	data frame, tibble, list, ...
...	Other arguments passed to the corresponding interfaced function.

Details

Interfaces call their corresponding interfaced function.

Value

Object returned by interfaced function.

Author(s)

Roberto Bertolusso

Examples

## Not run: 
library(intubate)
library(magrittr)
library(gss)


## ntbt_gssanova: Fitting Smoothing Spline ANOVA Models with Non-Gaussian Responses
data(bacteriuria)

## Original function to interface
gssanova(infect ~ trt + time, family="binomial", data = bacteriuria,
         id.basis = (1:820)[bacteriuria$id %in% c(3,38)], random = ~ 1 | id)
gssanova0(infect ~ trt + time, family="binomial", data = bacteriuria)
gssanova1(infect ~ trt + time, family="binomial", data = bacteriuria,
         id.basis = (1:820)[bacteriuria$id %in% c(3,38)], random = ~ 1 | id)

## The interface puts data as first parameter
ntbt_gssanova(bacteriuria, infect ~ trt + time, family="binomial",
              id.basis = (1:820)[bacteriuria$id %in% c(3,38)], random = ~ 1 | id)
ntbt_gssanova0(bacteriuria, infect ~ trt + time, family="binomial")
ntbt_gssanova1(bacteriuria, infect ~ trt + time, family="binomial",
               id.basis = (1:820)[bacteriuria$id %in% c(3,38)], random = ~ 1 | id)

## so it can be used easily in a pipeline.
bacteriuria %>%
  ntbt_gssanova(infect ~ trt + time, family="binomial",
                id.basis = (1:820)[bacteriuria$id %in% c(3,38)], random = ~ 1 | id)
bacteriuria %>%
  ntbt_gssanova0(infect ~ trt + time, family="binomial")
bacteriuria %>%
  ntbt_gssanova1(infect ~ trt + time, family="binomial",
                 id.basis = (1:820)[bacteriuria$id %in% c(3,38)], random = ~ 1 | id)


## ntbt_ssanova: Fitting Smoothing Spline ANOVA Models
data(nox)

## Original function to interface
ssanova(log10(nox) ~ comp*equi, data = nox)
ssanova0(log10(nox) ~ comp*equi, data = nox)

## The interface puts data as first parameter
ntbt_ssanova(nox, log10(nox) ~ comp*equi)
ntbt_ssanova0(nox, log10(nox) ~ comp*equi)

## so it can be used easily in a pipeline.
nox %>%
  ntbt_ssanova(log10(nox) ~ comp*equi)
nox %>%
  ntbt_ssanova0(log10(nox) ~ comp*equi)


## ntbt_ssanova9: Fitting Smoothing Spline ANOVA Models with Correlated Data
x <- runif(100); y <- 5 + 3*sin(2*pi*x) + rnorm(x)
dta <- data.frame(x, y)

## Original function to interface
ssanova9(y ~ x, data = dta, cov = list("arma", c(1, 0)))

## The interface puts data as first parameter
ntbt_ssanova9(dta, y ~ x, cov = list("arma", c(1, 0)))

## so it can be used easily in a pipeline.
dta %>%
  ntbt_ssanova9(y ~ x, cov = list("arma", c(1, 0)))


## ntbt_sscden: Estimating Conditional Probability Density Using Smoothing Splines
data(penny)

## Original function to interface
set.seed(5732)
sscden(~ year*mil, ~ mil, data = penny, ydomain = data.frame(mil=c(49, 61)))
sscden1(~ year*mil, ~ mil, data = penny, ydomain = data.frame(mil=c(49, 61)))

## The interface puts data as first parameter
set.seed(5732)
ntbt_sscden(penny, ~ year*mil, ~ mil, ydomain = data.frame(mil=c(49, 61)))
ntbt_sscden1(penny, ~ year*mil, ~ mil, ydomain = data.frame(mil=c(49, 61)))

## so it can be used easily in a pipeline.
set.seed(5732)
penny %>%
  ntbt_sscden(~ year*mil, ~ mil, ydomain = data.frame(mil=c(49, 61)))
penny %>%
  ntbt_sscden1(~ year*mil, ~ mil, ydomain = data.frame(mil=c(49, 61)))


## ntbt_sscox: Estimating Relative Risk Using Smoothing Splines
data(stan)

## Original function to interface
sscox(Surv(futime, status) ~ age, data = stan)

## The interface puts data as first parameter
ntbt_sscox(stan, Surv(futime, status) ~ age)

## so it can be used easily in a pipeline.
stan %>%
  ntbt_sscox(Surv(futime, status) ~ age)


## ntbt_ssden: Estimating Probability Density Using Smoothing Splines
data(aids)
## rectangular quadrature
quad.pt <- expand.grid(incu=((1:40)-.5)/40*100,infe=((1:40)-.5)/40*100)
quad.pt <- quad.pt[quad.pt$incu<=quad.pt$infe,]
quad.wt <- rep(1,nrow(quad.pt))
quad.wt[quad.pt$incu==quad.pt$infe] <- .5
quad.wt <- quad.wt/sum(quad.wt)*5e3

## Original function to interface
ssden(~ incu + infe, data = aids, subset = age >= 60,
      domain = data.frame(incu = c(0, 100), infe=c(0, 100)),
      quad = list(pt = quad.pt, wt = quad.wt))
ssden1(~ incu + infe, data = aids, subset = age >= 60,
       domain = data.frame(incu = c(0, 100), infe=c(0, 100)),
       quad = list(pt = quad.pt, wt = quad.wt))

## The interface puts data as first parameter
ntbt_ssden(aids, ~ incu + infe, subset = age >= 60,
           domain = data.frame(incu = c(0, 100), infe=c(0, 100)),
           quad = list(pt = quad.pt, wt = quad.wt))
ntbt_ssden1(aids, ~ incu + infe, subset = age >= 60,
            domain = data.frame(incu = c(0, 100), infe=c(0, 100)),
            quad = list(pt = quad.pt, wt = quad.wt))

## so it can be used easily in a pipeline.
aids %>%
  ntbt_ssden(~ incu + infe, subset = age >= 60,
             domain = data.frame(incu = c(0, 100), infe=c(0, 100)),
             quad = list(pt = quad.pt, wt = quad.wt))
aids %>%
  ntbt_ssden1(~ incu + infe, subset = age >= 60,
              domain = data.frame(incu = c(0, 100), infe=c(0, 100)),
              quad = list(pt = quad.pt, wt = quad.wt))


## ntbt_sshzd: Estimating Hazard Function Using Smoothing Splines
data(gastric)

## Original function to interface
sshzd(Surv(futime, status) ~ futime*trt, data = gastric)

## The interface puts data as first parameter
ntbt_sshzd(gastric, Surv(futime, status) ~ futime*trt)

## so it can be used easily in a pipeline.
gastric %>%
  ntbt_sshzd(Surv(futime, status) ~ futime*trt)


## ntbt_ssllrm: Fitting Smoothing Spline Log-Linear Regression Models
test <- function(x)
        {.3*(1e6*(x^11*(1-x)^6)+1e4*(x^3*(1-x)^10))-2}
x <- (0:100)/100
p <- 1-1/(1+exp(test(x)))
y <- rbinom(x,3,p)
y1 <- as.ordered(y)
y2 <- as.factor(rbinom(x,1,p))

dta <- data.frame(x, y1, y2)

## Original function to interface
ssllrm(~ y1*y2*x, ~ y1 + y2, data = dta)

## The interface puts data as first parameter
ntbt_ssllrm(dta, ~ y1*y2*x, ~ y1 + y2)

## so it can be used easily in a pipeline.
dta %>%
  ntbt_ssllrm(~ y1*y2*x, ~ y1 + y2)

## End(Not run)

intubate documentation built on May 2, 2019, 2:46 p.m.