Compute Cross-Covariance Decay Curves and Noise Floor

Evaluates window-matched cross-covariance matrices and noise-floor corrected Frobenius norms across multiple conversation subsets over a range of lag steps.

Usage

ena.ccd(
  x,
  codeNames = NULL,
  conversation_cols = NULL,
  max_window = 20,
  min_overlap = 10
)

Arguments

x

A data.frame, data.table, or an ENAdata/ena.set accumulation object.

codeNames

A character vector of column names corresponding to ENA codes.

conversation_cols

A character vector of column names defining discrete conversations.

max_window

An integer specifying the maximum lag to evaluate. Default is 20.

min_overlap

An integer specifying the minimum effective row overlap required per conversation at lag \(l\). Default is 10.

Value

An S3 object of class ena.ccd containing:

window_size

The estimated optimal window size (integer).

peak_lag

The lag corresponding to the peak cross-covariance norm.

curves

A data.frame of pooled cross-covariance metrics by lag.

codeNames

The evaluated code names.

conversation_cols

The conversation column names.

Examples

data(RS.data)
codeNames <- c("Data", "Technical.Constraints", "Performance.Parameters",
               "Client.and.Consultant.Requests", "Design.Reasoning", "Collaboration")
ccd_res <- ena.ccd(RS.data, codeNames = codeNames,
                   conversation_cols = c("Condition", "GroupName"))
print(ccd_res)
#> ENA Cross-Covariance Decay (CCD) Window Size Estimation
#> ------------------------------------------------------
#> Estimated Window Size : 6 
#> Peak Lag              : 1 
#> Codes Evaluated       : Data, Technical.Constraints, Performance.Parameters, Client.and.Consultant.Requests, Design.Reasoning, Collaboration 
#> Conversations By      : Condition, GroupName 
plot(ccd_res)