2 files changed, 187 insertions, 6 deletions
diff --git a/R/chemCal-package.R b/R/chemCal-package.R
index 8cc8c76..54ab190 100644
--- a/R/chemCal-package.R
+++ b/R/chemCal-package.R
@@ -54,7 +54,7 @@
 #' # LQ = 3.04 * LC (Currie 1999, p. 120)
 #' 3.04 * lod(m, alpha = 0.01, beta = 0.5)$x
 #' 
-NULL
+"din32645"
 
 
 
@@ -72,7 +72,7 @@ NULL
 #' Lewi, P.J., Smeyers-Verbeke, J. (1997) Handbook of Chemometrics and
 #' Qualimetrics: Part A, Chapter 8.
 #' @keywords datasets
-NULL
+"massart97ex1"
 
 
 
@@ -128,7 +128,7 @@ NULL
 #' # of 15, but the graphical procedure of Massart (p. 201) to derive the 
 #' # variances on which the weights are based is quite inaccurate anyway. 
 #' 
-NULL
+"massart97ex3"
 
 
 
@@ -147,7 +147,7 @@ NULL
 #' @source Rocke, David M. und Lorenzato, Stefan (1995) A two-component model
 #' for measurement error in analytical chemistry. Technometrics 37(2), 176-184.
 #' @keywords datasets
-NULL
+"rl95_cadmium"
 
 
 
@@ -167,7 +167,7 @@ NULL
 #' @source Rocke, David M. und Lorenzato, Stefan (1995) A two-component model
 #' for measurement error in analytical chemistry. Technometrics 37(2), 176-184.
 #' @keywords datasets
-NULL
+"rl95_toluene"
 
 
 
@@ -188,7 +188,81 @@ NULL
 #' Toronto.
 #' \url{https://sites.chem.utoronto.ca/chemistry/coursenotes/analsci/stats/index.html}
 #' @keywords datasets
-NULL
+"utstats14"
 
 
 
+
+
+#' Nitrite calibration data
+#' 
+#' Example dataset B.1 from DIN 38402 with concentrations in µg/L and the extinction
+#' as response measured using continuous flow analysis (CFA) according to 
+#' ISO 13395.
+#' 
+#' @name din38402b1
+#' @docType data
+#' @format A tibble containing 12 concentration levels with the respective 
+#' instrument response values.
+#' @references DIN 38402-51:2017-05, Beuth Verlag, Berlin. 
+#' https://dx.doi.org/10.31030/2657448
+#' @keywords datasets
+"din38402b1"
+
+
+
+
+
+#' Copper calibration data
+#' 
+#' Example dataset B.3 from DIN 38402. Cu was measured according to ISO 11885,
+#' using ICP-OES. The concentration are reported in mg/L and the response as 
+#' counts/s, describing the count of photons that are detected by the 
+#' photomultiplier detector of the device.
+#' 
+#' @name din38402b3
+#' @docType data
+#' @format A tibble containing 13 concentration levels and the respective 
+#' instrument response values.
+#' @references DIN 38402-51:2017-05, Beuth Verlag, Berlin. 
+#' https://dx.doi.org/10.31030/2657448
+#' @keywords datasets
+"din38402b3"
+
+
+
+
+
+#' Carbamazepin calibration data
+#' 
+#' Example dataset B.6 from DIN 38402 measured using LC-MS/MS. The 
+#' concentrations are reported in in µg/L and the response in arbitrary
+#' units (AU).
+#' 
+#' @name din38402b6
+#' @docType data
+#' @format A tibble containing 12 concentration levels and the respective 
+#' instrument response values.
+#' @references DIN 38402-51:2017-05, Beuth Verlag, Berlin. 
+#' https://dx.doi.org/10.31030/2657448
+#' @keywords datasets
+"din38402b6"
+
+
+
+
+
+#' Iron calibration data
+#' 
+#' Example dataset C.3 from DIN 38402 determined by ion chromatography.
+#' Concentrations are reported in mg/L and the extinction as response.
+#' 
+#' @name din38402c3
+#' @docType data
+#' @format A tibble containing 10 concentration levels and the respective 
+#' response values.
+#' @references DIN 38402-51:2017-05, Beuth Verlag, Berlin. 
+#' https://dx.doi.org/10.31030/2657448
+#' @keywords datasets
+"din38402c3"
+
diff --git a/R/linearity.R b/R/linearity.R
new file mode 100644
index 0000000..8a4f09b
--- /dev/null
+++ b/R/linearity.R
@@ -0,0 +1,107 @@
+#' Assess the linearity of a calibration curve
+#' 
+#' A function to create diagnostic plots for the assessment of the linearity of 
+#' calibration data based on their point-to-point slope or the curvature. 
+#' The underlying methods follow ISO 84 66-1:2021 and DIN 32 402-51:2017 
+#' (German Industrial Norm).
+#' 
+#' The point-to-point slope method is based on the assumption that the slope 
+#' between two points should not vary greatly within the linear range. 
+#' 
+#' The curvature method is similar to the point-to-point slope method. Here, 
+#' the ratio between the instrument signal and the concentration of the 
+#' calibration standard is assumed not to vary greatly within the linear range.
+#' 
+#' The use of the Mandel test is discouraged due to its limitations in the 
+#' identification of non-linear behaviour of calibration curves (Andrade and 
+#' Gomes-Carracedo, 2013). 
+#' 
+#' @param x numeric vector of independent values (usually concentrations).
+#' @param y numeric vector of dependent values (usually the signal of the 
+#' analytical device).
+#' @param method character string. Supported methods are "slope" and 
+#' "curvature".
+#' @param tolerance numeric value between 0 and 1, describing the acceptable
+#' deviation from the median of the slopes or the signal-to-concentration
+#' ratio. The default tolerance is 10%.
+#' @return returns a diagnostic plot 
+#' 
+#' @author Anıl Axel Tellbüscher
+#' 
+#' @importFrom graphics abline
+#' @importFrom graphics lines
+#' @importFrom stats median
+#' 
+#' @examples
+#' # Continuous Flow Analysis (CFA) data
+#' data(din38402b1)
+#' 
+#' # Point-to-point slope plot
+#' linearity(din38402b1$conc, din38402b1$ext, method = "slope")
+#' 
+#' # Curvature plot
+#' linearity(din38402b1$conc, din38402b1$ext, method = "curvature")
+#' 
+#' @references ISO 8466-1:2021. Water quality — Calibration and evaluation of 
+#' analytical methods — Part 1: Linear calibration function
+#' 
+#' J. M. Andrade and M. P. Gomez-Carracedo (2013) Notes on the use of 
+#' Mandel's test to check for nonlinearity in laboratory calibrations. 
+#' Analytical Methods 5(5), 1145 - 1149.
+#' 
+#' @export
+# Function to assess linearity of data using either slope or curvature method
+linearity <- function(x, y, method = c("slope", "curvature"), tolerance = 0.1) {
+  
+  # Check data integrity
+  # Ensure that x and y vectors have the same length
+  stopifnot("x and y must have the same length!" = length(x) == length(y))
+  
+  method <- match.arg(method)
+  
+  # Calculate the 'result' based on the chosen method
+  if (method == "slope") {
+    # For the 'slope' method, calculate the difference between consecutive points
+    x_diff = diff(x)        # Difference in x values
+    y_diff = diff(y)        # Difference in y values
+    result = y_diff / x_diff # Point-to-point slope (rate of change)
+  } else if (method == "curvature") {
+    # For the 'curvature' method, calculate the signal-to-concentration ratio
+    result = y / x # Element-wise division of y by x
+  }
+  
+  # Calculate the median of the results for tolerance check
+  result_median <- median(result)
+  
+  # Define upper and lower tolerance boundaries
+  upper_tolerance <- result_median + tolerance * result_median
+  lower_tolerance <- result_median - tolerance * result_median
+  
+  # Create a data frame to store the result and corresponding indices
+  df <- data.frame(result = result, index = 1:length(result))
+  
+  # Identify points that fall outside the tolerance range
+  outside_tolerance <- rbind(
+    subset(df, result > upper_tolerance), # Points above the upper tolerance
+    subset(df, result < lower_tolerance)  # Points below the lower tolerance
+  )
+  
+  # Basic scatter plot of the result against the index
+  plot(result ~ index, data = df, 
+       main = "linearity assessment", ylab = method, 
+       pch = 16)
+  
+  # Draw a line connecting all the points to visualize the trend
+  lines(df$index, df$result, col = "blue") # Blue line connecting points
+  
+  # Highlight points that are outside the tolerance range in red
+  points(x = outside_tolerance$index, y = outside_tolerance$result, 
+         pch = 16, col = "red")
+  
+  # Add a horizontal line at the median value of the result
+  abline(h = result_median, col = "red")
+  
+  # Add dashed horizontal lines at the upper and lower tolerance limits
+  abline(h = upper_tolerance, col = "red", lty = 3) # Upper tolerance
+  abline(h = lower_tolerance, col = "red", lty = 3) # Lower tolerance
+}