camel_index: Compute Multivariate-Weighted CAMEL Index

In CamelRatiosIndex: Multivariate-Weighted Indexing of CAMEL Ratios for Bank Performance

Compute Multivariate-Weighted CAMEL Index

Description

Computes a composite year-on-year index for bank performance assessment using the CAMEL framework. The multivariate weighting scheme employs robust factor analysis to derive communality-based weights from the correlation matrix of CAMEL ratios.

Usage

camel_index(
  base_data,
  current_data,
  bank_names = NULL,
  n_factors = 3,
  scale_data = TRUE,
  cov_control = rrcov::CovControlOgk(),
  method = "pca",
  scores_method = "regression"
)

Arguments

base_data	A data frame or matrix containing the base year CAMEL ratios. If a data frame, the first column must contain bank identifiers (character or numeric). If a matrix, bank identifiers must be supplied separately via bank_names.
current_data	A data frame or matrix containing the current year CAMEL ratios, in the same format and order as base_data.
bank_names	A character or numeric vector of bank identifiers. Required when base_data and current_data are matrices. Must be the same length as the number of rows in the data. Ignored when inputs are data frames.
n_factors	Integer specifying the number of factors to extract in the robust factor analysis. Default is 3.
scale_data	Logical indicating whether to standardize the data before factor analysis. Default is TRUE.
cov_control	A control object for robust covariance estimation, passed to robustfa::FaCov(). Default is rrcov::CovControlOgk().
method	Character specifying the factor analysis method. Default is "pca" (principal component analysis). See robustfa::FaCov() for options.
scores_method	Character specifying the method for computing factor scores. Default is "regression". See robustfa::FaCov() for options.

Details

The index is computed as the arithmetic mean of two multivariate-weighted Laspeyres-type and Paasche-type indices, scaled to a base of 100. The percentage difference (PD) from the base year is also reported.

Value

A list of class "camel_index" containing:

index_table

A tibble::tibble() with columns bank, I_mw (composite index, base = 100), and PD (percentage difference from base).

mw_lasp

Numeric vector of multivariate-weighted Laspeyres indices.

mw_pash

Numeric vector of multivariate-weighted Paasche indices.

I_mw

Numeric vector of composite indices (base = 100).

PD

Numeric vector of percentage differences from base year.

weights_base

Numeric vector of communality-based weights from base year factor analysis.

weights_current

Numeric vector of communality-based weights from current year factor analysis.

eigenvalues_base

Numeric vector of eigenvalues from base year correlation matrix.

eigenvalues_current

Numeric vector of eigenvalues from current year correlation matrix.

n_factors_base

Integer, number of eigenvalues > 1 in base year.

n_factors_current

Integer, number of eigenvalues > 1 in current year.

fa_base

The fitted robustfa::FaCov() object for base year.

fa_current

The fitted robustfa::FaCov() object for current year.

relativity_data

Matrix of current-to-base ratios for each CAMEL variable and bank.

base_data

The processed base year data (matrix, no bank names).

current_data

The processed current year data (matrix, no bank names).

bank_names

Character vector of bank identifiers.

n_factors

Integer, number of factors used.

call

The matched call.

Data Format

When supplying data frames, the first column must be the bank identifier (character or numeric), and the remaining columns must be the five CAMEL ratios in the standard order:

1. Capital Adequacy (Ca)

2. Asset Quality (Aq) – inverted internally

3. Management Efficiency (Me) – inverted internally

4. Earnings (Eq)

5. Liquidity (Lm) – inverted internally

The inversion of Aq, Me, and Lm is handled automatically because higher values of these ratios indicate worse bank performance.

Examples

# Using the built-in example data
base_year <- camel_2015
current_year <- camel_2022

result <- camel_index(base_year, current_year)
result$index_table

# Access individual components
result$mw_lasp
result$mw_pash
result$weights_base

# Using matrices with explicit bank names
base_mat <- as.matrix(camel_2015[, -1])
curr_mat <- as.matrix(camel_2022[, -1])
banks <- camel_2015$Bank

result2 <- camel_index(base_mat, curr_mat, bank_names = banks)
result2$index_table

CamelRatiosIndex documentation built on June 20, 2026, 5:07 p.m.