Introduction to seinfitR

Load Sample Data

We will use the glasshouse dataset included in the package:

library(seinfitR)

data(glasshouse, package = "seinfitR")

Fit the Seinhorst Model

We fit the model using the seinfitR() function, specifying initial parameter values and controlling iteration limits:

model <- seinfitR(
  p_i = "p_i", 
  y = "y", 
  data = glasshouse,
  start = list(m = 6, t = 6),
  control = seinfitR_control(maxiter = 20),
  z_fixed = TRUE
)
#> The Z_fixed parameter is set to TRUE: using the default value for z^t from Seinhorst (1986).
#> Model fitting completed successfully.

Model Result Details

The result of the seinfitR() function returns a list containing several important elements related to the fitted model:

- fit: The fitted model object, which can be further analyzed or extracted.
- summary_seinfitR: A summary of the fitted model, providing details about parameter estimates and statistical significance.
- cov: The covariance matrix of the parameter estimates, if available.
- data: The original dataset used for fitting the model.
- x: The name of the predictor variable (e.g., preplant nematode density).
- y: The name of the response variable (e.g., plant growth).
- z_fixed: A boolean indicating whether the `z` parameter was fixed during model fitting.
If `TRUE`, the function uses the default value for ( z^t ), as described in
Seinhorst, J. W. (1986). Effects of nematode attack on the growth and yield of crop plants. In Cyst nematodes (pp. 191-209). Springer US.

Model Summary

summary(model)
#> Seinhorst Model - Parameter Estimates
#> -----------------------------------------------------
#>         Estimate  Std. Error   t value     Pr(>|t|)
#> m      0.5951683 0.008177824  72.77832 4.096851e-16
#> t      1.6829177 0.116059892  14.50042 1.627030e-08
#> y_max 10.3675895 0.053752628 192.87596 9.127161e-21
#> -----------------------------------------------------
#> R2 - R squared (Coefficient of Determination):  0.9949782 
#> Adjusted_R2 - Adjusted R squared:               0.9940652 
#> -----------------------------------------------------

Extract Model Coefficients

coef(model)
#>          m          t      y_max 
#>  0.5951683  1.6829177 10.3675895

coef <- coef(model)

m <- coef[1]
t <- coef[2]
y_max <- coef[3]

cat("m:", m)
#> m: 0.5951683
cat("t:", t)
#> t: 1.682918
cat("y_max", y_max)
#> y_max 10.36759

Visualizing the Model Fit

The fitted model can be visualized using built-in plotting functions:

plot(model, main = "Seinhorst Model Fit for Glasshouse Data")

For further details, refer to the official GitHub repository.