The notes for the statistics lectures are in pdf format, and may be found here. The numbering of the pages is a little random, as they have been drawn in part from the notes for a previous class. They are in the correct order for this class in the pdf document, however.
Although there are problems listed below, you don't actually have to -do- them (although feel free to try them out!). Rather, you should look through them, think about them, and then look up and run the solutions. Likewise, you should run through the examples as well.
What you do have to do is a short weekly quiz, due on the following lab lecture day and which covers the material in the previous lecture. These quizzes are take-home and open book and notes, but are to be done individually and are covered by the honor system. They will count toward your final grade in lab.
The material contained here will be highly useful in the analysis of data in this lab, however we will obviously only get through all the material in class by the end of the semester. Thus, you will find it rewarding to read through the notes and look through the examples and problems as early as possible in the semester to extract what you can from the material. You will need it all again next fall, of course, when you take senior lab.
The dates for the lectures are: 2/18, 2/25, 3/11, 3/25, 4/8, 4/15, and 4/29, and they will be held in 136 Debartolo.
The following examples are demonstration Matlab codes which illustrate some of the concepts we will discuss in class
Example Index | |
---|---|
Example 1 | The Gaussian Distribution |
Example 2 | Error Propagation in Simple Functions |
Example 3, (energy.dat) | Statistical Analysis of the Covariance Matrix |
Example 4 | Matrix Solution to Linear Regression Problems |
Example 5 | Simple Error Propagation in Regression |
Example 5a, (miss.m) | Systematic Error in Regression |
Example 6 | More Sophisticated Error Analysis in Regression |
Example 7 | Error Estimation Using Undersampling and the Bootstrap |
Example 8 | More Systematic Error in Regression |
Example 9, (delta.m) | The Catalytic Oxidation of Methane |
The following problems (and solutions!) give you an idea of how simple Matlab programs can be written to analyze data and answer experimental questions such as "How many data points do I need to take?", and "How good is the answer I'm getting?", and "What is the data trying to tell me?"
Problem Index | ||
---|---|---|
Problem 1 | Estimation of Population Parameters | Solution |
Problem 2 | Error in Drag Calculations | Solution |
Problem 3 | Model Linearization and Error Propagation | Solution, (data.dat) |
Problem 4 | Error in Model Predictions | Solution |
Problem 5 | Weighted Linear Regression Analysis | Solution |
Problem 6 | The Bootstrap: Application to Heat Exchanger Analysis | Solution, (heat.dat) |
Problem 7 | Combined Linear and Non-Linear Regression | Solution, (problem7.dat, linfit.m) |
Due Date | Quiz | Solution |
---|---|---|
February 25 | Quiz 1 | Solution |
March 11 | Quiz 2 | Solution |
March 25 | Quiz 3 | Solution |
April 8 | Quiz 4 | Solution, (radgrad.m) |
April 15 | Quiz 5 | Solution |
April 29 | Quiz 6 | Solution (residfn.m) |