* How these regression equation are employed for prediction of yield and some other parameters on the basis of existing data sets ?
* How we developed non linear equation for prediction of phenology, growth and yield using existing datatsets of crop management ?
Non Linear Regression Modeling, its properties and example applications is discussed in details in the following book:
Applied Regression Analysis by Draper and Smith, Wiley Publications.
In Non linear regression modeling we start with a guess solution and iteratively try to converge to a solution satisfying some threshold criteria.
Hi, Yogesh,
An example: https://odajournal.com/2019/01/19/when-to-evaluate-a-nonlinear-model
You must first know the form of the equation that is suitable for your application. In some cases it comes from the solution of underlying differential equations (such as heat transfer or fluid flow). If not empirical relationships are often sued. If you think that a polynomial fit would be good for your system based on the way your data looks, then for simple one-variable systems Excel will suffice to come up with the regression coefficients and R-square value. If on the other hand some form of exponential or logarithm function is involved you can linearize the data first and try linear models. Again the form of the model must be known and programs like SAS can use PROC NLIN to solve for the unknown parameters (using Marquardt method for example).
Thank you so much for your help and references....it means a lot for me
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression