Dear participants,
Steps to find the powerpoint slides for a particular helping sessions1: Use the pop box to select the right day when session was delivered, Since only future day appear below except the first box, you may need to click the FIRST box to roll the calendar back to last month, remember our first class started on January 11th2: Click the date of your interesting event to expand the event of the day and look for the “More Details”, then you see the attached file3: Click the attached powerpoint file4: Click the event again, the expanded event box will return to a narrow line. I will post the slides right after each helping session.
Please click the course calendar to find our current tasks
Click here to find Info about your Instructor and tell me about yourself.
Best wish,
Sincerely
Hong Liu, Ph.D.
Mathematical Modeling and Simulation I 
December 19, 2013 at 10:43 am
Thank you for the great introduction and for helping to make this course available. I look forward to being part of this course and providing feedback in this forum.
Joseph-Ellison
December 19, 2013 at 2:54 pm
Hi Joseph
You’re most welcome. I hope you have a Merry Christmas.
See you soon
Hong LIU
January 30, 2017 at 2:05 pm
Dear all,
In order to collaborate in distance, I would like you to discuss many topics relevant to this course, including the reflection questions in each lesson, homework, or project problems. The blogs that you post will be score in the end as your participation grade up to 10% of your final grade. There are a few exceptions that I will clearly tell you not to discuss, e.g. take-home test, individual conceptual modeling, etc.
You must first enter the course website, then, either start a new thread or follow a thread based on the topics that you are interested to discuss. It does not have to be formal, it can be a few sentences of casual chat. You may just test a few lines of your inputs and see what is happen on the website.
Best wish,
Dr. Hong Liu
January 30, 2017 at 6:13 pm
Reflection Questions (Solving Linear Equations):
1) You can represent the equations of lines and planes without expilictly naming all of the variables if you use a matrix for the information instead of equations.
2) To find out if a line will intersect with a planar region, you can set the two equations equal to one another and solve for the variables. If the solution is true then at those variable values the line and plane will intersect.(I don’t know how to do this if you don’t know the equation for the line or plane though. Advice please)
3) OpenGL stands for Open Graphics Library, this is a cross platform programming interface that allows the user to render 2D and 3D graphs/plots. But I can’t find the correct Matlab commands to complete the two tasks above. (If anybody knows those commands please let me know, thanks)
-Sarah Terrell
Math Modeling
Adams State University
January 30, 2017 at 8:14 pm
This is exact what I am expecting and hope some others will follow and you learn from each other. You also share the knowledge that you find from wiki and internet. My interest is how the machine learning can also grade such a work instead of grading by the instructors.
Dr. Hong Liu
January 31, 2017 at 2:13 am
Reflection Questions (Linear Transformation)
1. The address can be translated to CTRS and CIRS coordinates using transformation matrices. The transformation matrices for CIRS to CTRS is the inverse of the transformation matrix for CTRS to CIRS.
2. The CIRS does not rotate with the Earth, so there will be different CIRS coordinates in different hours of the day.
3. I would assume the GPS receiver translates the position with transformation matrices that are respective to the coordinate systems being used, however I am not certain of this.
Rebecca Harvey
Embry-Riddle Aeronautical University
February 1, 2017 at 2:13 pm
You can find very nice wiki page and simulation website by typing the keywords. Please try the relevant exercise in Hw 3 and discuss what you find.
Thanks for your blog.
March 3, 2017 at 10:25 pm
Reflection Questions Linear Regression:
1) To figure out what type of curve we should select to fit our data points we should use the formula for linear regression. Or we should figure out a way to use matlab or some program to find the line of best fit using linear regression.
2) We justify that a particular curve of the selected type best fits the data points by using confidence intervals of 95% or 99% and making sure that the majority of points falls within that given interval. Our line needs to be within that confidence interval. We then need to evaluate the confidence intervals(how large or small is the variance?) and evaluate the standard deviation of the curve.
3) We evaluate the quality of the data by seeing how many data points, the more there are the better the estimation will be. And we also need to check for any outliners in the data set. If our data set is mainly between[50,100]x[20,30] then if we have a point at (20,3) then we know there is a possibility of a measurement error and may want to exclude this point from the observation of data.
-Sarah Terrell
Adams State
March 10, 2017 at 10:22 pm
Group Project:
Since we don’t have reflection questions right now, I’ll make a post about the project.
I am looking forward to working on the group project. This will give us all a great chance to practice some real world applications. I assume all of us at some point or another will need to solve a real world ODE or PDE in whatever field we end up in. This will be a good experience for us all, having to work in a group to produce research. It will be interesting to learn Stella and other programs that we might end up learning through this process. Hopefully these are applications which still have relevance in the work place as well.
-Sarah Terrell
Adams State University
March 11, 2017 at 2:49 pm
Sarah,
I will work two more classes with you on Monday and Wednesday from your regular class time to make up the two classes about modeling methodology And classification. I will assign you the hw 1 for the interesting problems. In your spring break, you may miss the three hours in Stella modeling. I suggest your team choose the discrete models so you only need to use NetLogo, which free you from the trouble of Stella.