This assignment developed learning on: variance-covariance matrix, diagonalizing a matrix, eigenvectors and eigen values, and the practical application of how eigen values can be used to identify principal components (features that contribute the most to the identification of a data analytics problem). More specifically, how Principal Component Analysis (PCA) can be used to reduce the dimensions of data in a data analytics problem.
The assignment is to write a program that includes all the functionality described in four pans shown in Figure 1 below.
Step A is to generate two different samples, from a normal distribution, where the sample size is two for the first sample, and 100 for the second sample. The Mean of those two samples is given as a vector and the covariance matrix is given as a matrix.
Step B plots the generated samples.
Write a
“Teislatilif normal distribution specified by N N.100 31,;,…1
E l4 :1.
IX Let io.!s pis,..nd +0r1 0.53,4, IMmonalize these Mo matrices.
suiaciattseinatrixxesosglixd. Figure 1 Assignment 1 Step C computes the sample mean and the sample covariance matrix. Step D provides two matrices and asks that the matrices be diagonalized.
You are required to undertake this assessment as an individual. Each student will work on the new consumer product ‘Stan Australia’ that is suitable for review of the effectiveness of its product launch and commercialisation. You need to identify a close competitor to the new product. Your review needs to be undertaken from the perspective of the company introducing the new product and the company producing the ‘old’ product (a close competitor of Stan). In conducting the review, you may examine the product in both the on and off-line environment. It is expected that you will undertake both primary research (which will involve observations of retailing of the product and of consumer consumption behaviour), and secondary research to identify any reviews of the products, or past products in the category, in the media
March 30, 2015
6:16 pm
Read More »
