symptoms: Symptoms data

In randomLCA: Random Effects Latent Class Analysis

Symptoms data

Description

This is the data for Beath and Heller (2009).

Allergy and respiratory symptoms for infants 0 to 2 years in six month periods. Outcome is presence or absence of symptom in the six months. Original data was collected at Visits 1-7 over the 2 year period which were summarised to six month periods.

Note that these models can be slow to fit, with the "symptoms.lca2random2" model taking about 1-2 hours.

Thanks to the investigators of the CAPS study for making the data available.

Usage

symptoms

Format

A data frame with 444 observations on the following 17 variables.

Nightcough.13

Night cough in visits 1-3

Wheeze.13

Wheeze in visits 1-3

Itchyrash.13

Itchy rash in visits 1-3

FlexDerma.13

Flexural Dermatitis in visits 1-3

Nightcough.45

Night cough in visits 1-3

Wheeze.45

Wheeze in visits 4-5

Itchyrash.45

Itchy rash in visits 4-5

FlexDerma.45

Flexural Dermatitis in visits 4-5

Nightcough.6

Night cough in visit 6

Wheeze.6

Wheeze in visit 6

Itchyrash.6

Itchy rash in visit 6

FlexDerma.6

Flexural Dermatitis in visits 1-3

Nightcough.7

Night cough in visit 7

Wheeze.7

Wheeze in visit 7

Itchyrash.7

Itchy rash in visit 7

FlexDerma.7

Flexural Dermatitis in visit 7

Freq

Number of subjects

Source

Mihrshai et al (2001)

References

Mihrshahi, S., Peat, J.K., Webb, K., Tovey, R.E., Marks, G.B., Mellis, C.M. and Leeder S.R. (2001) The Childhood Asthma Prevention Study (CAPS): Design and research protocol of a randomized trial for the primary prevention of asthma. Control led Clinical Trials, 22:333–354.

Beath, K.J. and Heller, G.Z. (2009) Latent trajectory modelling of multivariate binary data. Statistical Modelling, 9(3):199–213.

Examples

symptoms.lca2 <- randomLCA(symptoms[, 1:16], freq = symptoms$Freq, nclass = 2,
  cores = 1)
symptoms.lca2random <- randomLCA(symptoms[, 1:16], freq = symptoms$Freq, 
	random = TRUE, nclass = 2, blocksize = 4, constload = FALSE, cores = 1)
symptoms.lca2random2 <- randomLCA(symptoms[, 1:16], freq = symptoms$Freq, 
	random = TRUE, level2 = TRUE, nclass = 2, level2size = 4, constload = FALSE,
	penalty = 0.1, cores = 1)

randomLCA documentation built on Sept. 24, 2024, 1:06 a.m.