This function helps to fit a Mitscherlich-style exponential response model for relative yield (ry) as a function of soil test values (stv).
Usage
mits_formula_1(x, a, b, c)
mits_formula_2(x, b, c)
mits_formula_3(x, c)
mitscherlich(
data = NULL,
stv,
ry,
type = 1,
target = 95,
tidy = TRUE,
plot = FALSE,
resid = FALSE
)
boot_mitscherlich(data, stv, ry, type = 1, n = 999, target = 95, ...)
Arguments
- x
selfstart vector. for model fit Default: NULL
- a
selfstart arg. for asymptote parameter, Default: NULL
- b
selfstart arg. for b parameter (b = -X_intercept) Default: NULL
- c
selfstart arg. for curvature parameter Default: NULL
- data
Optional argument to call and object of type data.frame or data.table containing the stv and ry data, Default: NULL
- stv
name of the vector containing soil test values (-) of type
numeric
.- ry
name of the vector containing relative yield values (%) of type
numeric
.- type
string or number that indicates the type of Mitscherlich model to fit. Default: 1. For model with 'no restrictions' use
type = 1
,type = "no restriction"
, ortype = "free"
; For model with 'asymptote = 100' usetype = 2
,type = "asymptote 100"
, ortype = "100"
; For model with 'asymptote = 100 and xintercept = 0'"type = 3
,type = "asymptote 100 from 0"
, ortype = "fixed"
.- target
numeric
value of relative yield target (e.g. 90 for 90%) to estimate the CSTV. Default: NULL- tidy
logical operator (TRUE/FALSE) to decide the type of return. TRUE returns a tidy data frame or tibble (default), FALSE returns a list.
- plot
logical operator (TRUE/FALSE) to plot the Mitscherlich model, Default: FALSE
- resid
logical operator (TRUE/FALSE) to plot residuals analysis, Default: FALSE
- n
sample size for the bootstrapping Default: 500
- ...
when running bootstrapped samples, the
...
(open arguments) allows to add grouping variable/s (factor or character) Default: NULL
Value
returns an object of type ggplot
if plot = TRUE.
returns a residuals plot if resid = TRUE.
returns an object of class data.frame
if tidy = TRUE,
returns an object of class list
if tidy = FALSE.
Mitscherlich type 1 formula
Mitscherlich type 2 formula
Mitscherlich type 3 formula
mitscherlich: function
boot_mitscherlich: bootstrapping function
Details
See online-documentation for additional details.
Note
For extended reference, we recommend to visit: https://github.com/austinwpearce/SoilTestCocaCola by Austin Pearce.
References
Melsted, S.W. and Peck, T.R. (1977). The Mitscherlich-Bray Growth Function. In Soil Testing (eds T. Peck, J. Cope and D. Whitney). doi:10.2134/asaspecpub29.c1
Examples
# \donttest{
# Example dataset
dat <- data.frame("ry" = c(65,80,85,88,90,94,93,96,97,95,98,100,99,99,100),
"stv" = c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15))
# Run
fit_example_mits <- mitscherlich(data = dat, type = 1,
ry = ry, stv = stv, resid = TRUE, plot = FALSE)
fit_example_mits
#> # A tibble: 1 × 13
#> asymptote b curvature equation y_intercept target CSTV AIC AICc BIC
#> <dbl> <dbl> <dbl> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 98.7 2.07 0.371 98.7(1-e… 52.8 95 6.77 64.0 68.0 66.8
#> # ℹ 3 more variables: R2 <dbl>, RMSE <dbl>, pvalue <dbl>
# }