本次澳洲作业主要为大数据统计相关的统计代写assignment
问题 1
令 X 为具有多元正态分布 X N( ; V ) 的 p 维随机向量
其中 是 p 维向量，V 是非奇异 p p 对称矩阵。
(a) [1] 矩阵非奇异是什么意思？
(b) [1] 如果给你一个矩阵，你如何检查它是非奇异的？
(c) [1] 举一个只有 f0 的 3×3 矩阵的例子； 1g-条目是单数。
(d) [1] 举一个只有 f0 的 3×3 矩阵的例子； 1g 条目是非奇异的。
(e) [1] 比较这两个矩阵的特征值。你能说什么
奇异矩阵的特征值？
(f) [3] 证明 Z (X )W 1 具有 N(0; I) 分布，其中 I 是 p p 恒等式
矩阵，W 是满足 W2 = V 的对称矩阵。
问题2
考虑 p = 768 维向量和 p p 协方差矩阵包含在
文件“mu.txt”和“Sigma.txt”。将这些文件中的值读入 R。
(a) [1] 来自带参数的多元正态分布的样本 n = 1 个实例
和。将 p = 768 维样本重塑为大小为 r c = 28 28 的矩阵
并绘制矩阵。你能猜出你正在生成什么随机图像吗？你
可能需要采样/尝试几次才能正确猜测。
(b) [1] 计算的特征值并将它们绘制为线图。在什么值下
特征值似乎趋于平稳？即，“肘部”出现的值是多少。
Question 1
Let X be a p-dimensional random vector with multivariate Normal distribution X N(; V )
where is a p-dimensional vector and V is a non-singular p p symmetric matrix.
(a) [1] What does it mean for a matrix to be non-singular?
(b) [1] If you are given a matrix how could you check that it is non-singular?
(c) [1] Come up with an example of a 3×3 matrix with only f0; 1g-entries that is singular.
(d) [1] Come up with an example of a 3×3 matrix with only f0; 1g-entries that is non-singular.
(e) [1] Compare the eigenvalues of these two matrices. What can you say about the
eigenvalues of the singular matrix?
(f) [3] Show that Z (X )W 1 has a N(0; I) distribution where I is a p p identity
matrix and W is a symmetric matrix satisfying W2 = V .
Question 2
Consider the p = 768 dimensional vector and p p covariance matrix contained in
the files ‘mu.txt’ and ‘Sigma.txt’. Read the values from these files into R.
(a) [1] Sample n = 1 instance from a multivariate Normal distribution with parameters
and . Reshape the p = 768 dimensional sample to a matrix of size r c = 2828
and plot the matrix. Can you guess what random images you are generating? You
may need to sample/try this a couple of times to guess correctly.
(b) [1] Calculate the eigenvalues of and plot them as a line plot. At what value do the
eigenvalues seem to level off at? i.e., what is the value where the “elbow” appears.