Given a data matrix X as in Question 2, Assume the means of the p variables...
Let x1, x2 denote the variables for the two-dimensional data summarized by the covariance matrix, the eigenvalues, and the unit eigenvectors shown below. Find a new variable y, of the form y 1 +ewh2-1, such that y, has maximum possible variance over the given data. How much of the variance in the data is explained by y,? 74.968-14.032. λι 78.245, 0.23 -0.97 u,-| and λ,-14.891, u,- 0.23 14.032 18.168 -0.97
Let x1, x2 denote the variables for the two-dimensional data...
Let X=(X1,…,Xn)′ be the n×p data matrix, where Xi=(Xi1,…,Xip)′ is the ith observation. Let X¯=n−1∑ni=1Xi be the sample mean. Let sj1j2=1/n∑ni=1(Xij1−X¯j1)(Xij2−X¯j2) be the sample covariance between the j1th and j2th variables. Let S=(sj1j2) be the sample covariance matrix. Show that S=1nX′X−X¯′X¯.