bootsem: Bootstrap Analysis of the Fitted System of Structural...

In BigSEM: Constructing Large Systems of Structural Equations

Description

Use the bootstrap data sets to evaluate the significance of regulatory effects in the fitted large system of structural equations by the two-stage penalized least squares (2SPLS) method proposed by Chen, Zhang and Zhang (2016).

Usage

bootsem(y,x,sk,nboots=100)

Arguments

y	a data frame containing the endogenous variables Y_1, Y_2, ..., Y_p in the model.
x	a data frame containing the exogenous variables X_1, X_2, ..., X_q in the model.
sk	a list with the k-th element specifying S_k which includes the indices of exogenous variables appearing in the structural equation for k-th endogenous variable.
nboots	number of the bootstrap datasets.

Details

Generate the bootstrap data sets by randomly sampling from the original data with replacement, and apply 2SPLS to each bootstrap data set to infer the regulatory effects.

Value

boot.target	the index of target variable.
boot.source	the index of source variable.
boot.freq	the bootstrap frequency of the regulatory effect.
boot.mean	the mean of the regulatory effect.
boot.sd	the standard deviation of the regulatory effects.

Author(s)

Chen Chen <[e-mail]>, Dabao Zhang <[e-mail]>.

References

Chen, C., Zhang, M., and Zhang, D. (2016) A Two-Stage Penalized Least Squares Method for Constructing Large Systems of Structural Equations. (Submitted)

See Also

fitsem for constructing large systems of structural equations.

Examples

data(simdata)
attach(simdata)
#simsem <- fitsem(y=y,x=x,sk=sk)
#btres <- bootsem(y,x,sk,nboots=200)
#summary(btres)

Example output

BigSEM documentation built on May 2, 2019, 3:37 p.m.