Q3 Regression Consider the lasso criterion as seen in the lectures L = YTY - BTXTY + ½ß³ XTXẞ + \||B||1. In this problem, you will consider the case with only one explanatory variable. The columns X and Y are centered so there is no intercept in the model. The parameter ẞ is a scalar and so are the quantities XTX,XTY and YTY. These scalars have values XTX = 2.5, XTY = -0.5 and YTY = 2.5. A) Write the ordinary least squares estimate B = B) Using the given quantities and the result of B, write the lasso criterion as a function of ẞß and λ (both scalars). The criterion is L = By doing derivative of L and solving the corresponding system, determine the lasso coefficient as a function of A. That is Â¹ = Â¹ (A)= D) The value of A at which the path ẞL (A) shrinks to zero is

Q3 Regression Consider the lasso criterion as seen in the lectures L = YTY - BTXTY + ½ß³ XTXẞ + \||B||1. In this problem, you will consider the case with only one explanatory variable. The columns X and Y are centered so there is no intercept in the model. The parameter ẞ is a scalar and so are the quantities XTX,XTY and YTY. These scalars have values XTX = 2.5, XTY = -0.5 and YTY = 2.5. A) Write the ordinary least squares estimate B = B) Using the given quantities and the result of B, write the lasso criterion as a function of ẞß and λ (both scalars). The criterion is L = By doing derivative of L and solving the corresponding system, determine the lasso coefficient as a function of A. That is Â¹ = Â¹ (A)= D) The value of A at which the path ẞL (A) shrinks to zero is

Algebra & Trigonometry with Analytic Geometry

13th Edition

ISBN:9781133382119

Author:Swokowski

Publisher:Swokowski

Chapter3: Functions And Graphs

Section3.3: Lines

Problem 76E

See similar textbooks