1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How should I teach myself Linear Algebra?

  1. Sep 1, 2015 #1
    I am preparing myself for a BSEE starting next year. I have just finished Stewarts Calculus for Calc I,II,III and Diff Eqs up to solving 2nd order non-hom diff eqs by undet. coeffs, variation of parameters and power series.

    In high school some 15 years ago I saw some very basic linear algebra in advanced math; matrix addition, multiplication, inverse, Gauss-Jordan Elimination and I kind of understand what determinants are.

    About 7 years ago I took a Linear Algebra unit during my Chemistry degree, and understood very little. Saliently, I had no clue what all that Eigen-stuff was, and still only have a little bit of intuition about it from some scattered reading and discussion.

    I have looked at three options for teaching myself Linear Algebra over the next 6 months

    1) work through Axler's Linear Algebra Done Right - though I have read that this is a second course book meant for undergrad math majors or post-grad students.

    2) The MIT OCW unit on Linear Algebra: http://ocw.mit.edu/courses/mathematics/18-06-linear-algebra-spring-2010/

    This uses Introduction to Linear Algebra by Gilbert Strang

    3) work through a physical text Elementary Linear Algebra: Application Version by Anton and Rorres

    The idea of using the physical text has some appeal to me. And this text seems quite comparable to MIT's Intro to Linear Algebra text from Strang.

    What would your advice be?

    edit: I am also quite familiar with Matlab.
  2. jcsd
  3. Sep 1, 2015 #2
    If you want to go through a physical text, why not use Strang's book?
    I would recommend working through the OCW, and doing extra problems from the text. I think there are section of the book that are not covered in the course. You can do those as well.
    At the same time, in your spare time, work large, real-world problems with MATLAB. This is what I wish I had done. For example, you can use eigenvector decomposition to do image compression. And then there is the ever-popular stock market predictor.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook