EGR 103/Spring 2016/Lab 8

The following document is meant as an outline of what is covered in this assignment.

Typographical Errors

None yet!

Part of the goal here is to see the work that you can offload to a function (taking the A, T, and Order and producing the P, Sr, r2, AMin, and TMin) and then using it for a variety of orders.
The other part is to see that while, mathematically, "more is better," having a higher order may not make much of an improvement. The key here is to figure out the "just right" order where adding terms make important changes in how the model predicts the values.
You can use polyfit and polyval to get the coefficients and estimates.
Don't forget to have the \(S_t\) value somewhere!

This problem is a straight-forward general linear regression.
If you want, you can use the General Linear Regression Example Code or you can start from scratch. If you use the example, be sure to use the appropriate version of the honor code.

This is a multidimensional general linear fit.
If you want, you can use the General Multidimensional Linear Regression Example Code or you can start from scratch. If you use the example, be sure to use the appropriate version of the honor code.
Don't forget to calculate the required estimate and error in that estimate!

These are both general linear regressions.
If you want, you can use the General Linear Regression Example Code or you can start from scratch. If you use the example, be sure to use the appropriate version of the honor code.
Be sure you create two graphs for each piece - one for the total population and one with the three populations shown separately.
Also be sure to answer all the questions.

This problem uses a linearized version of power law to solve for the coefficients.
If you want, you can use the Linearized Regression Example Code or you can start from scratch. If you use the example, be sure to use the appropriate version of the honor code.
You do not need to include the extra code for the other models if you want to make the program shorter.

This is a nonlinear fit.
If you want, you can use the Nonlinear Regression Example Code or you can start from scratch. If you use the example, be sure to use the appropriate version of the honor code.
Don't forget to justify your choice of initial conditions!

You will be looking at both a linearized method and nonlinear regression in this problem.
Be sure to name the variables that relate to the different models in a way that you can easily keep track of them.
Make sure that the graph you turn in is the graph from the original data set - not the altered versions.
You are not required to include graphs of the altered versions, but you may want to discuss them in your report.
Be sure to include all six models. There is an empty table waiting for you.