Data Science in Theory and Practice. Maria Cristina Mariani. Читать онлайн. Newlib. NEWLIB.NET

Автор: Maria Cristina Mariani
Издательство: John Wiley & Sons Limited
Серия:
Жанр произведения: Математика
Год издания: 0
isbn: 9781119674733
Скачать книгу
as

bold sigma-summation equals cov left-parenthesis bold upper X right-parenthesis equals Start 4 By 4 Matrix 1st Row 1st Column sigma Subscript 1 comma 1 Baseline 2nd Column sigma Subscript 1 comma 2 Baseline 3rd Column midline-horizontal-ellipsis 4th Column sigma Subscript 1 comma p Baseline 2nd Row 1st Column sigma Subscript 2 comma 1 Baseline 2nd Column sigma Subscript 2 comma 2 Baseline 3rd Column midline-horizontal-ellipsis 4th Column sigma Subscript 2 comma p Baseline 3rd Row 1st Column vertical-ellipsis 2nd Column vertical-ellipsis 3rd Column Blank 4th Column vertical-ellipsis 4th Row 1st Column sigma Subscript p comma 1 Baseline 2nd Column sigma Subscript p comma 2 Baseline 3rd Column midline-horizontal-ellipsis 4th Column sigma Subscript p comma p Baseline EndMatrix period

      Let u equals bold b Superscript upper T Baseline bold upper X, where bold b is a vector of constants different from bold a. The population covariance of z equals bold a Superscript upper T Baseline bold upper X and u equals bold b Superscript upper T Baseline bold upper X is defined as

cov left-parenthesis z comma u right-parenthesis equals cov left-parenthesis bold a Superscript upper T Baseline bold upper X comma bold b Superscript upper T Baseline bold upper X right-parenthesis equals sigma Subscript z u Baseline equals bold a Superscript upper T Baseline sigma-summation bold b comma

      where sigma-summation denotes the population covariance matrix which is defined in (3.5).

StartLayout 1st Row 1st Column rho Subscript z u 2nd Column equals corr left-parenthesis bold a Superscript upper T Baseline bold upper X comma bold b Superscript upper T Baseline bold upper X right-parenthesis 2nd Row 1st Column Blank 2nd Column equals StartFraction sigma Subscript z u Baseline Over StartRoot sigma Subscript z Superscript 2 Baseline EndRoot StartRoot sigma Subscript u Superscript 2 Baseline EndRoot EndFraction 3rd Row 1st Column Blank 2nd Column equals StartFraction bold a Superscript upper T Baseline sigma-summation bold b Over StartRoot bold a Superscript upper T Baseline sigma-summation bold a EndRoot StartRoot bold b Superscript upper T Baseline sigma-summation bold b EndRoot EndFraction comma EndLayout

      where sigma-summation denotes the population covariance matrix which is defined in (3.5).

      Remarks 3.1 If bold upper A is a scalar matrix and bold upper X are random vectors, then bold AX represents several linear combinations. The population mean vector and covariance matrix are given by

upper E left-parenthesis bold AX right-parenthesis equals bold upper A upper E left-parenthesis bold upper X right-parenthesis equals bold upper A mu comma cov left-parenthesis bold AX right-parenthesis equals bold upper A sigma-summation bold upper A Superscript upper T Baseline comma

      where mu denotes the population mean vector and sigma-summation denotes the population covariance matrix which is defined in (3.5).

      The proof of Remark 3.1 is left as an exercise. Please see Problem 2.

      Please refer to Johnson and Wichern (2014), Rencher (2002), and Axler (2002) and references therein for more details of multivariate analysis.

      1 The following are four measurements on the variables and :91251600143587Use the above information to answer the following questions:Find the sample mean vector, .Find the sample covariance matrix, .Find the sample correlation matrix, .

      2 For random vectors and , scalar matrices and , and scalar vectors and , prove the following:..., where denotes the population covariance matrix which is defined in (3.5)...

      3 Consider the five pairs of measurements :3426855.54410CalculateThe sample means and .The sample variances and .The sample covariances

      4 Consider the data matrix:Calculate the matrix of deviations (residuals), . Is this matrix of full rank?Note: A matrix is of full rank if all rows and columns are linearly independent. A square matrix is full rank if and only if its determinant is nonzero.

      5 Calculate the sample covariance matrix using the data matrix in Problem 4.

      6 Consider the data matrix:Obtain the mean corrected data matrix, sample covariance matrix and verify that the columns are linearly dependent.

      7 If for , show that , where is a constant.

      8 If for , show that , where is a constant.

      9 The data in Table 3.1 (Elston and Grizzle 1962) consist of measurements , and of the ramus bone at four different ages on each of 20 boys.Find .Find .Find .Table 3.1 Ramus Bone Length at Four Ages for 20 Boys.AgeIndividual147.848.849.049.7246.447.347.748.4346.346.847.848.5445.145.346.147.2547.648.548.949.3652.553.253.353.7751.253.054.354.5849.850.050.352.7948.150.852.354.41045.047.047.348.31151.251.451.651.91248.549.253.055.51352.152.853.755.01448.248.949.349.81549.650.451.251.81650.751.752.753.31747.247.748.449.51853.354.655.155.31946.247.548.148.42046.347.651.351.8

      10 For the data in Table 3.1, define and .Find , , , and using (3.12) and (3.13).Find and using (3.14) and (3.15).

      Конец ознакомительного фрагмента.

      Текст предоставлен ООО «ЛитРес».

      Прочитайте эту книгу целиком, купив