Graduate Course Proposal Form Submission Detail - OCE6609
Edit function not enabled for this course.
Approved by SCNS
Submission Type: New
Course Change Information (for course changes only):
Comments: Elective - Marine Science. To GC. Note same number as Data Program. Pending confirmation of numbers. Approved; To USF Sys 4/21/16; to SCNS after 4/28/16. SCNS approved 6888 as 6809 eff 7/1/16
- Department and Contact Information
Tracking Number Date & Time Submitted 5249 2015-07-27 Department College Budget Account Number Marine Science MS 250000 Contact Person Phone Don Chambers 7275533351 firstname.lastname@example.org
- Course Information
Prefix Number Full Title OCE 6609 Data Analysis Methods Is the course title variable? N Is a permit required for registration? N Are the credit hours variable? N Is this course repeatable? N If repeatable, how many times? 0 Credit Hours Section Type Grading Option 3 C - Class Lecture (Primarily) R - Regular Abbreviated Title (30 characters maximum) Data Analysis Course Online? Percentage Online - 0
This course introduces students to common statistical techniques like linear regression, Fourier series, low-pass filtering, optimal interpolation, and principal component analysis that are commonly used to analyze time-series and mapped data.
A. Please briefly explain why it is necessary and/or desirable to add this course.
Replacing Selected Topics with Permanent number; already listed in program
B. What is the need or demand for this course? (Indicate if this course is part of a required sequence in the major.) What other programs would this course service?
Students from the College of Marine Science take this as their option for learning data analysis.
C. Has this course been offered as Selected Topics/Experimental Topics course? If yes, how many times?
Yes, 3 or more times
D. What qualifications for training and/or experience are necessary to teach this course? (List minimum qualifications for the instructor.)
Terminal Degree Required
- Other Course Information
The objective of the class is to give students basic knowledge for applying statistical methods to analyzing time-series and mapped data, to better quantify uncertainty in the analyzed data, and to gain insight into how some data they use are prepared.
B. Learning Outcomes
Student outcomes will be learning the derivation of the algorithms used for data analysis, learning to program the algorithms, and learning how to apply them to various data sets.
C. Major Topics
Basic statistics (mean, standard deviation, correlation, covariance)
Linear regression using ordinary least squares
Uncertainty, including Monte Carlo testing assuming random residuals
Fourier Series and applications to computing spectra
Sampling theory and aliasing
Smoothing/filtering a time-series
Uncertainty analysis assuming a colored (correlated) noise model
Mapping data to grids
Principal component analysis
E. Course Readings, Online Resources, and Other Purchases
To do the labs, we will use a freely available software package that is designed to work on multiple platforms. Make sure you download the appropriate version for your operating system (Mac, Windows, Linux) and follow the on-line instructions for installing it.
Make sure to also download the “Introduction to SciLab” file from here:
as it’s a good reference.
F. Student Expectations/Requirements and Grading Policy
Grades will be based on completing 5 practical lab assignments that use methods presented in the class. Only students who consistently turn in assignments by the date given will be able to earn an A or A+. The actual grade will be determined by the content of the assignment, and how well I judge that you understand the concepts. No student who turns in assignments consistently will earn less than a B, even if the assignment is not fully complete or is not done correctly. The reason I want you to turn in assignments on time, even if you feel you have not completed it, is so that I can give you comments/suggestions on how to improve your analysis skills. This will definitely help you on later assignments. Any student who consistently turns in assignments more than a week late (without prior permission from me), or does not turn in assignments at all will automatically receive a grade of C.
Analysis of data is more than just obtaining statistics, but also to use these numbers in a narrative to reach a scientific conclusion. I will grade you both on obtaining the proper statistics as well as how you express your results. The more descriptive and complete your analysis is, the higher your grade will be, provided the answers are correct. For instance, a grade in the range of 97-100 (A+) will have figures/tables with appropriate labels and captions, as well as a short but descriptive narrative that refers to the figures/tables appropriately, and all answers will be correct. A grade in the range of 87-90 (B+) will be given for assignments that may have correct answers, but poor narrative.
I have designed the assignments so that they should not be terribly time consuming. Many can be mostly completed except for the write-up during our classroom lab periods. Data and basic software for the lab assignments will be provided, and example problems will be demonstrated in class. Part of the goal of this class is to teach students to write (or adapt) their own processing routines, and so parts of the labs will entail doing this. However, there will be plenty of time in class with me and other students to help with any issues you may have, so don't worry about it too much. The ultimate goals of the labs are to familiarize the student with the steps required to analyze a particular data set, obtain more insight into the effects of different filter parameterizations, and to better understand quantifying error.
G. Assignments, Exams and Tests
Specific Schedule of Lectures and Labs
January 5, Lecture # 1, Overview and philosophy of class; review of basic statistics; Autocovariance/Autocorrelation.
January 7, Lab # 1, Intro to SciLab; reading in data; plotting; computing statistics including autocovariance.
January 12, Lab class to understanding basic statistics, plotting, functions, scripting in SciLab.
January 14, Lecture # 2, Review of linear algebra; introduction to ordinary least squares estimation (fitting a trend; least square regression)
January 19, MLK day, no class
January 21, Lecture # 3, Uncertainty analysis for least squares assuming random residuals; Monte Carlo simulations; residual analysis. Calculating degrees of freedom.
January 26, Lab # 2 Least squares estimation; fitting trends to data in presence of other periodic variations; analysis of residuals; estimating uncertainty assuming random residuals based on Degrees of Freedom. HW # 1 (due by 2/13).
February 28, Lab # 2 (cont.), running a Monte Carlo simulation.
February 2,4: No class, I am out of town
February 9, Lecture # 4, Fourier series analysis and periodograms
February 11, Lab # 3, Computing Fourier Series from time-series and plotting periodogram. HW # 2 (due by 2/27).
February 16, Lab # 4, intro to Fast Fourier Transforms in SciLab and how they are like/unlike Fourier Series (no assignment, in class only).
February 18, time in class to answer questions on Fourier Series and FFTs that have come up while working on problems.
February 23, Lecture # 5, Sampling theory & aliasing.
February 25, Lecture # 6, Smoothing/Filtering using a weighted average.
March 2-6, Spring Break, no classes
March 9, Lecture # 7, Smoothing/Filtering (cont.) and using to create colored noise to improve uncertainty estimate in least squares.
March 11, Lab # 5; Smoothing a time-series using a weighted average. HW # 3 (due by 3/27).
March 16, Lab # 5 (cont); using smoothing to create colored noise for Monte Carlo
March 18, Lecture # 8, Intro to optimal interpolation, using a time-series (incl. uncertainty)
March 23, Lab # 6, using optimal interpolation to fill gaps in data (no assignment, in class).
March 25, No class, I am out of town.
March 30, Lecture # 9, Introduction to mapping data & grids
April 1, Lecture # 10, Mapping with weighted averages
April 6, Lab # 7, Mapping in SciLab and plotting results. HW # 4 (due by 4/17).
April 8, Lab # 7 (cont), in class work on mapping.
April 13, Lecture # 11, Principal component analysis & empirical orthogonal functions
April 15, Lab # 8, Using SVD in SciLab to perform an EOF analysis of a set of tide gauges. HW # 5 (due by 4/24).
April 20, Lecture # 12,
April 22, Lecture # 13, More on spatial optimal interpolation
H. Attendance Policy
Course Attendance at First Class Meeting – Policy for Graduate Students: For structured courses, 6000 and above, the College/Campus Dean will set the first-day class attendance requirement. Check with the College for specific information. This policy is not applicable to courses in the following categories: Educational Outreach, Open University (TV), FEEDS Program, Community Experiential Learning (CEL), Cooperative Education Training, and courses that do not have regularly scheduled meeting days/times (such as, directed reading/research or study, individual research, thesis, dissertation, internship, practica, etc.). Students are responsible for dropping undesired courses in these categories by the 5th day of classes to avoid fee liability and academic penalty. (See USF Regulation – Registration - 4.0101,
Attendance Policy for the Observance of Religious Days by Students: In accordance with Sections 1006.53 and 1001.74(10)(g) Florida Statutes and Board of Governors Regulation 6C-6.0115, the University of South Florida (University/USF) has established the following policy regarding religious observances: (http://usfweb2.usf.edu/usfgc/gc_pp/acadaf/gc10-045.htm)
In the event of an emergency, it may be necessary for USF to suspend normal operations. During this time, USF may opt to continue delivery of instruction through methods that include but are not limited to: Blackboard, Elluminate, Skype, and email messaging and/or an alternate schedule. It’s the responsibility of the student to monitor Blackboard site for each class for course specific communication, and the main USF, College, and department websites, emails, and MoBull messages for important general information.
I. Policy on Make-up Work
Attendance Policy: While class attendance is not mandatory, students are responsible for all material presented in lectures. Due dates for lab exercises are given at the beginning of the semester, and the lab write-ups are expected to be turned in on time. If a student does not turn in the exercise on time and does not have prior permission from the instructor, 5 points will be deducted from the grade for each day it is late, to a maximum of 50 points.
Student Conduct, Rights and Responsibilities, and Academic Integrity: Please read page 7 of the 2014-5 CMS graduate Student Handbook, which can be found at: http://www.marine.usf.edu/documents/handbook-2014-2015.pdf
J. Program This Course Supports
- Course Concurrence Information