derived: Data on blood pressure body mass and age
In ggm: Graphical Markov Models with Mixed Graphs

Description Usage Format References Examples

Raw data on blood pressure, body mass and age on 44 female patients, and covariance matrix for derived variables.

1	data(derived)

A list containing a dataframe raw with 44 lines and 5 columns and a symmetric 4x4 covariance matrix S.

The following is the description of the variables in the dataframe raw

Sys: Systolic blood pressure, in mm Hg
Dia: Diastolic blood pressure, in mm Hg
Age: Age of the patient, in years
Hei: Height, in cm
Wei: Weight, in kg

The following is the description of the variables for the covariance matrix S.

Y: Derived variable Y=log(Sys/Dia)
X: Derived variables X=log(Dia)
Z: Body mass index Z=Wei/(Hei/100)^2
W: Age

Wermuth N. and Cox D.R. (1995). Derived variables calculated from similar joint responses: some characteristics and examples. Computational Statistics and Data Analysis, 19, 223-234.

# A DAG model with a latent variable U
G = DAG(Y ~ Z + U, X ~ U + W, Z ~ W)

data(derived)

# The model fitted using the derived variables
out = fitDagLatent(G, derived$S, n = 44, latent = "U")

# An ancestral graph model marginalizing over U
H = AG(G, M = "U")

# The ancestral graph model fitted obtaining the 
# same result
out2 = fitAncestralGraph(H, derived$S, n = 44)