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
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
|
---
output: github_document
---
<!-- README.md is generated from README.rmd. Please edit that file -->
# chemCal - Calibration functions for analytical chemistry
<!-- badges: start -->
[](https://cran.r-project.org/package=chemCal)
[](https://app.travis-ci.com/github/jranke/chemCal)
[](https://codecov.io/github/jranke/chemCal)
<!-- badges: end -->
## Overview
chemCal is an R package providing some basic functions for conveniently working
with linear calibration curves with one explanatory variable.
## Installation
From within [R][r-project], get the official chemCal release using
```r
install.packages("chemCal")
```
## Usage
chemCal works with univariate linear models of class `lm`. Working with one of
the datasets coming with chemCal, we can produce a calibration plot using the
`calplot` function:
### Plotting a calibration
```r
library(chemCal)
m0 <- lm(y ~ x, data = massart97ex3)
calplot(m0)
```
<!-- -->
### LOD and LOQ
If you use unweighted regression, as in the above example, we can calculate a
Limit Of Detection (LOD) from the calibration data.
```r
lod(m0)
#> $x
#> [1] 5.407085
#>
#> $y
#> [1] 13.63911
```
This is the minimum detectable value (German: Erfassungsgrenze), i.e. the
value where the probability that the signal is not detected although the
analyte is present is below a specified error tolerance beta (default is 0.05
following the IUPAC recommendation).
You can also calculate the decision limit (German: Nachweisgrenze), i.e.
the value that is significantly different from the blank signal
with an error tolerance alpha (default is 0.05, again following
IUPAC recommendations) by setting beta to 0.5.
```r
lod(m0, beta = 0.5)
#> $x
#> [1] 2.720388
#>
#> $y
#> [1] 8.314841
```
Furthermore, you can calculate the Limit Of Quantification (LOQ), being
defined as the value where the relative error of the quantification given the
calibration model reaches a prespecified value (default is 1/3).
```r
loq(m0)
#> $x
#> [1] 9.627349
#>
#> $y
#> [1] 22.00246
```
### Confidence intervals for measured values
Finally, you can get a confidence interval for the values
measured using the calibration curve, i.e. for the inverse
predictions using the function `inverse.predict`.
```r
inverse.predict(m0, 90)
#> $Prediction
#> [1] 43.93983
#>
#> $`Standard Error`
#> [1] 1.576985
#>
#> $Confidence
#> [1] 3.230307
#>
#> $`Confidence Limits`
#> [1] 40.70952 47.17014
```
If you have replicate measurements of the same sample,
you can also give a vector of numbers.
```r
inverse.predict(m0, c(91, 89, 87, 93, 90))
#> $Prediction
#> [1] 43.93983
#>
#> $`Standard Error`
#> [1] 0.796884
#>
#> $Confidence
#> [1] 1.632343
#>
#> $`Confidence Limits`
#> [1] 42.30749 45.57217
```
## Reference
You can use the R help system to view documentation, or you can
have a look at the [online documentation][pd-site].
[r-project]: https://www.r-project.org/
[pd-site]: https://pkgdown.jrwb.de/chemCal/
|
