Package 'surveynnet'

Simple body fat example data

Description

Simple body fat example data

Usage

body_fat

Format

body_fat

A data frame with 12 rows and 9 columns:

Subject

Subject ID

group

group

Weight_kg

weight

Height_cm

height

Age

age

pct_body_fat

percent body fat

survey_wt

survey weight

stratum

stratum

cluster

cluster

---

Nhanes example

Description

Nhanes example

Usage

nhanes.demo

Format

nhanes.demo

A data frame with 6230 rows and 9 columns:

SEQN

Respondent sequence number

SDMVPSU

Masked variance pseudo-PSU

SDMVSTRA

Masked variance pseudo-stratum

WTMEC2YR

Full sample 2 year MEC exam weight

BMXHT

Standing height (cm)

BMXWT

Weight (kg)

BMXBMI

Body maxx index (kg/m**2)

BPXSY1

Systolic blood pressure

BPXDI1

Diastolic blood pressure

---

Predict response from fitted nnet, using new data

Description

Predict response from fitted nnet, using new data

Usage

## S3 method for class 'surveynnet'
predict(object, newdat, flag = 1, ...)

Arguments

object	The surveynnet object (returned by surveynnet())
newdat	The matrix or data frame of test examples. Must be of the same structure as the data matrix used to fit the surveynnet object.
flag	This argument determines which of the three fitted models are used to predict the response for the newdat data. Valid entries are 1,2, or 3: 1: the CSD model 2: the weighted model 3: the base model
...	Additional arguments to be passed in. NOTE, currently this function does not accept additional arguments.

Value

The matrix/vector of values returned by the trained network.

Examples

# From the example in `surveynnet` help file:
y <- body_fat$pct_body_fat
x <- body_fat[,c("Weight_kg", "Height_cm", "Age")]
weight <- body_fat$survey_wt
strat <- body_fat$stratum
clust <- body_fat$cluster
y[strat==1] <- y[strat==1] + 30*0.00015*rnorm(sum(strat==1))
y[strat==2] <- y[strat==2] + 30*0.15*rnorm(sum(strat==2))
myout <- surveynnet(x,y,weight = weight, strat = strat, clust=clust)
newdat <-  2*x+rnorm(dim(x)[1])
predict(object = myout, newdat = newdat, flag = 1)

---

Neural Net for Complex Survey Data

Description

The surveynnet package extends the functionality of nnet (Venables and Ripley, 2002), which already supports survey weights, by enabling it to handle clustered and stratified data. It achieves this by incorporating design effects through the use of effective sample sizes in the calculations, performed by the package described in Valliant et al. (2023), by following the methods outlined by Chen and Rust (2017) and Valliant et al. (2018).

Usage

surveynnet(
  x,
  y,
  weight = 1,
  strat = NULL,
  clust = NULL,
  comp_cases = FALSE,
  wtnorm = FALSE,
  a = 1,
  s = 1,
  clnest = FALSE,
  ...
)

Arguments

x	Matrix or data frame of predictors. Must not contain any missing values.
y	Vector of targets / response values. Must not contain any missing values.
weight	The weights for each sample.
strat	The stratum for each sample.
clust	The cluster for each sample.
comp_cases	If TRUE, filter out missing values from x, y, weight, strat, and clust. Default FALSE. Note that in either case, the dimensions of all data mentioned above must agree.
wtnorm	If TRUE, weights are normalized prior to fitting both weighted and design-effect-adjusted weighted models
a, s	These are arguments for the softplus scaling function used to stabilize the design-effect adjustment. The softplus scaling function is f(x,a,s) = 1 + log(1 + e^{s*(x - a)})
clnest	Are cluster IDs numbered within strata (TRUE or FALSE, default = FALSE)? If TRUE, cluster IDs can be restarted within strata, e.g., 1,2,3,1,2,3,... This is passed into PracTools::deffCR
...	Additional arguments to be passed into PracTools::deffCR or nnet::nnet. See documentation of those packages and functions for more details. Note that for the neural net (nnet), the default here is set to 3 layers ("size" parameter) and maximum iterations ("maxit" parameter) is set to 2000.

Value

A list containing four objects:

• A dataframe with the fitted values of the neural nets, using: no weights ("fitted"), the user-inputted weights ("fitted_weighted"), and the new method that adjusts the weights by using a design effect incorporating cluster and strata ("fitted_deff").

• Each of the three fitted neural network objects (base model, weighted model, and the novel CSD model); this can be used to predict the outcomes for new observations.

References

• Chen, S., and K. F. Rust. 2017."An Extension of Kish’s Formula for Design Effects to Two- and Three-Stage Designs with Stratification.”, Journal of Survey Statistics and Methodology,5 (2): 111–30.

• Valliant, R., J. A. Dever, and F. Kreuter. 2018. Practical Tools for Designing and Weighting Survey Samples .2nd ed. New York: Springer-Verlag.

See Also

PracTools::deffCR() which this function wraps.

Examples

# short example with body fat dataset
y <- body_fat$pct_body_fat
x <- body_fat[,c("Weight_kg", "Height_cm", "Age")]
weight <- body_fat$survey_wt
strat <- body_fat$stratum
clust <- body_fat$cluster
y[strat==1] <- y[strat==1] + 30*0.00015*rnorm(sum(strat==1))
y[strat==2] <- y[strat==2] + 30*0.15*rnorm(sum(strat==2))

myout <- surveynnet(x,y,weight = weight, strat = strat, clust=clust)
myout


# NHANES example
# Predicting Diastolic BP from BMI, Systolic BP and Height
# PLEASE NOTE: for this example, pass "clnest=TRUE"

x <- nhanes.demo[,c("BMXBMI", "BPXSY1", "BMXHT")]
weight <- nhanes.demo$WTMEC2YR
strat <- nhanes.demo$SDMVSTRA
clust <- nhanes.demo$SDMVPSU
y <- nhanes.demo$BPXDI1
myout <- surveynnet(x,y,weight = weight, strat = strat, clust=clust, clnest=TRUE)
head(myout$results, 15)

---