Scribd
Upload
2K views4 pages

Intermediate R - Nonlinear Regression in R

Nonlinear regression is an extension of linear regression that allows for modeling nonlinear relationships between variables. It uses a nonlinear mathematical model rather than a linear one. Parameter estimation in nonlinear regression uses iterative procedures to estimate parameters, which may require supplying initial parameter estimates. Examples of nonlinear regression models included in the document are the logistic growth curve model and other models like exponential, monomolecular, Gompertz, and Weibull.

Uploaded by

Vivay Salazar
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
2K views4 pages

Intermediate R - Nonlinear Regression in R

Nonlinear regression is an extension of linear regression that allows for modeling nonlinear relationships between variables. It uses a nonlinear mathematical model rather than a linear one. Parameter estimation in nonlinear regression uses iterative procedures to estimate parameters, which may require supplying initial parameter estimates. Examples of nonlinear regression models included in the document are the logistic growth curve model and other models like exponential, monomolecular, Gompertz, and Weibull.

Uploaded by

Vivay Salazar
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 4

Nonlinear Regression

• An extended linear regression technique


Nonlinear Regression in R in which a nonlinear mathematical model
is used to describe the relationship
between the response variable and the
predictor variables.
• A nonlinear regression model is a model
Violeta I. Bartolome that contains at least one of the
Senior Associate Scientist
parameters in a nonlinear form
PBGB-CRIL
v.bartolome@cgiar.org • Example

Y = αXβ + ε ε ~ N( 0, σ 2 )

Linearized Nonlinear Model Parameter Estimation


• The linearized form of the example is: • Iterative procedures are used
log Y = log α + β log X • May require users to supply initial
• The normality assumption for the error term estimates of parameters
will not be appropriate for the linearized • User should know the biological
form. interpretation of the parameters to
• Hence, if a nonlinear model is assumed to provide intelligent initial values.
be appropriate for the data, the nonlinear • The function getInitial() may also be
model should be used for the analysis used.
instead of the linearized form.
Read and Plot Data
Logistic growth curve
• Sample data
Graph shows a logistic
Severity represents the number
of lesions on the leaf surface,
relationship.
standardized as a proportion of
the maximum. Logistic model:

α
y=
1 + eβ −rt

Initial values Estimate Parameters

SSlogis creates initial estimates of the parameters Asym,


xmid, and scal from the model:
Asym
y= xmid−input xmid
1+e scal α = Asym β=
scal
The model that we are estimating is: r = 1 t = input
scal
α
y=
1 + eβ −rt
Check fit of data

Other Non-linear Models Other Non-linear Models


• Exponential • Monomolecular

y = y0erx + ε y = 1 − (1 − y0 )e −rx + ε
Other Non-linear Models Other Non-linear Models
• Logistic • Gompertz

1
y=
1 − y0 y = e log y0 e −rx
1+
y0e −rx

Other Non-linear Models


• Weibull

c
 x −a 
− 
y = 1−e  b  Thank you!

You might also like