Furthermore the log likelihood is convex with respect to β and this is true for

Furthermore the log likelihood is convex with respect

This preview shows page 43 - 46 out of 46 pages.

Furthermore, the log likelihood is convex with respect to β , and this is true for all φ ( x ), so all of the methods proceed identically. COS424/SML 302 Linear Regression February 25, 2019 43 / 46
Image of page 43

Subscribe to view the full document.

Regression with non-linear relationships Polynomial functions We might consider the model: y i = φ ( x i ) T β + φ ( x ) = [1 , x , x 2 , . . . , x p ] T This allows us to find non-linear (polynomial) relationships between x , y . Higher-order interactions between predictors We can include non-additive interactions, e.g., for x i = [ x 1 , x 2 ] i : φ ( x ) = [ x 1 , x 2 , x 1 x 2 , x 2 1 x 2 , . . . ] Note that we consider interactions among predictors, not among samples. COS424/SML 302 Linear Regression February 25, 2019 44 / 46
Image of page 44
Summary A few key points from this discussion: Regression models, like other models, must be built thoughtfully: What are the set of predictors to include? Should I include an intercept term? Should I include any non-linear terms? When possible, try the normal equation to fit the ˆ β coefficients With large numbers of samples, online methods might be faster than batch methods With large numbers of predictors, these methods may be unstable Optimization methods are an important tool in the ML toolbox We will discuss regression over the next three lectures. COS424/SML 302 Linear Regression February 25, 2019 45 / 46
Image of page 45

Subscribe to view the full document.

Additional resources Regression is a standard tool in statistics; resources abound MLAPA: Chapter 7 Elements of Statistical Learning , Chapter 4 Metacademy: Linear Regression COS424/SML 302 Linear Regression February 25, 2019 46 / 46
Image of page 46

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern

Ask Expert Tutors You can ask 0 bonus questions You can ask 0 questions (0 expire soon) You can ask 0 questions (will expire )
Answers in as fast as 15 minutes