Description Usage Arguments Details Examples
View source: R/repeatAnalysis.r
Derive the proportion of similarities in the shape of event traces between sensors within two sessions.
1 2 3 | repeatAnalysis(x, sensorToMatch = 4, sensorToMatchAgainst = 3:5,
whichJSONLabel = "pfmc3x5s_rest30s", sessionOrder = NULL, B = 1000,
toPlot = FALSE)
|
x |
An "FemFit" object. |
sensorToMatch |
A numeric vector which denotes sensors to use in the comparison from |
sensorToMatchAgainst |
A numeric vector which denotes sensors to use in the comparison from |
whichJSONLabel |
A character argument specifying part of the protocol to extract the event traces from. |
sessionOrder |
A character vector of length two specifying the sessions to compare. Defaults to |
B |
A numeric argument specifying the number of bootstraps to run to estimate the variability of the propotion of similarities. |
toPlot |
A logical argument as to whether |
repeatAnalysis()
first standardises the event traces against themselves, procuring the raw shape of the event trace.
Then, it uses loess regression to get a smooth approximation of the shape for each event trace.
A total of (n_1 + n_2)/2 observations are drawn sequentially over time from each loess regression to construct the smooth approximation, where n_x is the number of observations that make up the xth event trace.
A vector of differences is constructed and then squared. A Hidden Markov Model is fitted to the vector of differences assuming two states which adhere to a Normal distribution.
Finally, the proportion of similarity is derived by using the output of the Viterbi algorithm, and the assessment of the variability is done with a parametric bootstrap based on the Viterbi algorithm's delta probabilities.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 | AS005 = read_FemFit(c(
"Datasets_AukRepeat/61aa0782289af385_283_csv.zip",
"Datasets_AukRepeat/61aa0782289af385_284_csv.zip"
),
remove.NAs = TRUE
) %>%
# Segment the FemFit data
segment(
cp. = 0.001,
numOfNodesToLabel = list(c(3, 1, 3, 4), c(4, 1, 5, 3))
) %>%
# Calculate the zeroed out pressures
deriveZero(method = "lmm")
# Produce the proportions of similarities for Sensor four against Sensor four with the ggplot2 objects.
AS005_RepeatOutput = repeatAnalysis(AS005, 4, 4, "pfmc3x5s_rest30s", toPlot = TRUE)
# View the proportion of similarities for the three events, with the bootstrapped standard errors and CIs
AS005_RepeatOutput %>%
select(eventID, classProb, classProb_stderr, classProb_lwrbnd, classProb_uprbnd)
# View the ggplot2 objects generated by `repeatAnalysis()`
AS005_RepeatOutput$plotObj[1]
AS005_RepeatOutput$plotObj[2]
AS005_RepeatOutput$plotObj[3]
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.