lars/FYS-STK4155
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
270b0beff9b052d3cf776769c8f3eb18c05c6ea9
FYS-STK4155/doc/pub/week37/html/week37.html
T
mhjensen 397e904837 updating html and ipynbs
2020-09-16 13:58:39 +02:00

1114 lines
56 KiB
HTML
Raw Blame History

<!--
Automatically generated HTML file from DocOnce source
(https://github.com/hplgit/doconce/)
-->
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="DocOnce: https://github.com/hplgit/doconce/" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<meta name="description" content="Week 37: Ridge and Lasso Regression">
<title>Week 37: Ridge and Lasso Regression</title>
<style type="text/css">
/* bloodish style */
body {
font-family: Helvetica, Verdana, Arial, Sans-serif;
color: #404040;
background: #ffffff;
}
h1 { font-size: 1.8em; color: #8A0808; }
h2 { font-size: 1.6em; color: #8A0808; }
h3 { font-size: 1.4em; color: #8A0808; }
h4 { color: #8A0808; }
a { color: #8A0808; text-decoration:none; }
tt { font-family: "Courier New", Courier; }
/* pre style removed because it will interfer with pygments */
p { text-indent: 0px; }
hr { border: 0; width: 80%; border-bottom: 1px solid #aaa}
p.caption { width: 80%; font-style: normal; text-align: left; }
hr.figure { border: 0; width: 80%; border-bottom: 1px solid #aaa}
.alert-text-small { font-size: 80%; }
.alert-text-large { font-size: 130%; }
.alert-text-normal { font-size: 90%; }
.alert {
padding:8px 35px 8px 14px; margin-bottom:18px;
text-shadow:0 1px 0 rgba(255,255,255,0.5);
border:1px solid #bababa;
border-radius: 4px;
-webkit-border-radius: 4px;
-moz-border-radius: 4px;
color: #555;
background-color: #f8f8f8;
background-position: 10px 5px;
background-repeat: no-repeat;
background-size: 38px;
padding-left: 55px;
width: 75%;
}
.alert-block {padding-top:14px; padding-bottom:14px}
.alert-block > p, .alert-block > ul {margin-bottom:1em}
.alert li {margin-top: 1em}
.alert-block p+p {margin-top:5px}
.alert-notice { background-image: url(https://cdn.rawgit.com/hplgit/doconce/master/bundled/html_images/small_gray_notice.png); }
.alert-summary { background-image:url(https://cdn.rawgit.com/hplgit/doconce/master/bundled/html_images/small_gray_summary.png); }
.alert-warning { background-image: url(https://cdn.rawgit.com/hplgit/doconce/master/bundled/html_images/small_gray_warning.png); }
.alert-question {background-image:url(https://cdn.rawgit.com/hplgit/doconce/master/bundled/html_images/small_gray_question.png); }
div { text-align: justify; text-justify: inter-word; }
</style>
</head>
<!-- tocinfo
{'highest level': 2,
'sections': [('Plans for week 37', 2, None, '___sec0'),
('Thursday September 10', 2, None, '___sec1'),
('A Bayesian approach to develop intuition about skrinkage '
'methods',
2,
None,
'___sec2'),
('The singular value decomposition', 2, None, '___sec3'),
('Linear Regression Problems', 2, None, '___sec4'),
('Fixing the singularity', 2, None, '___sec5'),
('Basic math of the SVD', 2, None, '___sec6'),
('The SVD, a Fantastic Algorithm', 2, None, '___sec7'),
('Economy-size SVD', 2, None, '___sec8'),
('Codes for the SVD', 2, None, '___sec9'),
('Mathematical Properties', 2, None, '___sec10'),
('Friday September 12', 2, None, '___sec11'),
('Ridge and LASSO Regression', 2, None, '___sec12'),
('More on Ridge Regression', 2, None, '___sec13'),
('Interpreting the Ridge results', 2, None, '___sec14'),
('More interpretations', 2, None, '___sec15'),
('A better understanding of regularization', 2, None, '___sec16'),
('Decomposing the OLS and Ridge expressions',
2,
None,
'___sec17'),
('Introducing the Covariance and Correlation functions',
2,
None,
'___sec18'),
('Correlation Function and Design/Feature Matrix',
2,
None,
'___sec19'),
('Covariance Matrix Examples', 2, None, '___sec20'),
('Correlation Matrix', 2, None, '___sec21'),
('Correlation Matrix with Pandas', 2, None, '___sec22'),
('Correlation Matrix with Pandas and the Franke function',
2,
None,
'___sec23'),
('Rewriting the Covariance and/or Correlation Matrix',
2,
None,
'___sec24'),
('Linking with SVD', 2, None, '___sec25')]}
end of tocinfo -->
<body>
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
TeX: {
equationNumbers: { autoNumber: "AMS" },
extensions: ["AMSmath.js", "AMSsymbols.js", "autobold.js", "color.js"]
}
});
</script>
<script type="text/javascript" async
src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML">
</script>
<!-- ------------------- main content ---------------------- -->
<center><h1>Week 37: Ridge and Lasso Regression</h1></center> <!-- document title -->
<p>
<!-- author(s): Morten Hjorth-Jensen -->
<center>
<b>Morten Hjorth-Jensen</b> [1, 2]
</center>
<p>
<!-- institution(s) -->
<center>[1] <b>Department of Physics, University of Oslo</b></center>
<center>[2] <b>Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University</b></center>
<br>
<p>
<center><h4>Sep 16, 2020</h4></center> <!-- date -->
<br>
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec0">Plans for week 37 </h2>
<ul>
<li> Thursday September 10: Motivation for shrinkage methods and the Singular Value Decompostion theorem</li>
<li> Friday September 11: Ridge and Lasso regression</li>
</ul>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec1">Thursday September 10 </h2>
<p>
<a href="https://www.uio.no/studier/emner/matnat/fys/FYS-STK4155/h20/forelesningsvideoer/LectureSeptember10.mp4?vrtx=view-as-webpage" target="_blank">Video of Lecture</a> and <a href="https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/NotesSeptember10.pdf" target="_blank">handwritten notes</a>
<p>
More material will be added here, see handwritten notes also.
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec2">A Bayesian approach to develop intuition about skrinkage methods </h2>
<p>
See Handwritten notes
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec3">The singular value decomposition </h2>
<p>
<div class="alert alert-block alert-block alert-text-normal">
<b></b>
<p>
<p>
The examples we have looked at so far are cases where we normally can
invert the matrix \( \boldsymbol{X}^T\boldsymbol{X} \). Using a polynomial expansion as we
did both for the masses and the fitting of the equation of state,
leads to row vectors of the design matrix which are essentially
orthogonal due to the polynomial character of our model. Obtaining the inverse of the design matrix is then often done via a so-called LU, QR or Cholesky decomposition.
<p>
This may
however not the be case in general and a standard matrix inversion
algorithm based on say LU, QR or Cholesky decomposition may lead to singularities. We will see examples of this below.
<p>
There is however a way to partially circumvent this problem and also gain some insights about the ordinary least squares approach, and later shrinkage methods like Ridge and Lasso regressions.
<p>
This is given by the <b>Singular Value Decomposition</b> algorithm, perhaps
the most powerful linear algebra algorithm. Let us look at a
different example where we may have problems with the standard matrix
inversion algorithm. Thereafter we dive into the math of the SVD.
</div>
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec4">Linear Regression Problems </h2>
<p>
One of the typical problems we encounter with linear regression, in particular
when the matrix \( \boldsymbol{X} \) (our so-called design matrix) is high-dimensional,
are problems with near singular or singular matrices. The column vectors of \( \boldsymbol{X} \)
may be linearly dependent, normally referred to as super-collinearity.
This means that the matrix may be rank deficient and it is basically impossible to
to model the data using linear regression. As an example, consider the matrix
$$
\begin{align*}
\mathbf{X} & = \left[
\begin{array}{rrr}
1 & -1 & 2
\\
1 & 0 & 1
\\
1 & 2 & -1
\\
1 & 1 & 0
\end{array} \right]
\end{align*}
$$
<p>
The columns of \( \boldsymbol{X} \) are linearly dependent. We see this easily since the
the first column is the row-wise sum of the other two columns. The rank (more correct,
the column rank) of a matrix is the dimension of the space spanned by the
column vectors. Hence, the rank of \( \mathbf{X} \) is equal to the number
of linearly independent columns. In this particular case the matrix has rank 2.
<p>
Super-collinearity of an \( (n \times p) \)-dimensional design matrix \( \mathbf{X} \) implies
that the inverse of the matrix \( \boldsymbol{X}^T\boldsymbol{X} \) (the matrix we need to invert to solve the linear regression equations) is non-invertible. If we have a square matrix that does not have an inverse, we say this matrix singular. The example here demonstrates this
$$
\begin{align*}
\boldsymbol{X} & = \left[
\begin{array}{rr}
1 & -1
\\
1 & -1
\end{array} \right].
\end{align*}
$$
We see easily that \( \mbox{det}(\boldsymbol{X}) = x_{11} x_{22} - x_{12} x_{21} = 1 \times (-1) - 1 \times (-1) = 0 \). Hence, \( \mathbf{X} \) is singular and its inverse is undefined.
This is equivalent to saying that the matrix \( \boldsymbol{X} \) has at least an eigenvalue which is zero.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec5">Fixing the singularity </h2>
<p>
If our design matrix \( \boldsymbol{X} \) which enters the linear regression problem
$$
\begin{align}
\boldsymbol{\beta} & = (\boldsymbol{X}^{T} \boldsymbol{X})^{-1} \boldsymbol{X}^{T} \boldsymbol{y},
\label{_auto1}
\end{align}
$$
has linearly dependent column vectors, we will not be able to compute the inverse
of \( \boldsymbol{X}^T\boldsymbol{X} \) and we cannot find the parameters (estimators) \( \beta_i \).
The estimators are only well-defined if \( (\boldsymbol{X}^{T}\boldsymbol{X})^{-1} \) exits.
This is more likely to happen when the matrix \( \boldsymbol{X} \) is high-dimensional. In this case it is likely to encounter a situation where
the regression parameters \( \beta_i \) cannot be estimated.
<p>
A cheap <em>ad hoc</em> approach is simply to add a small diagonal component to the matrix to invert, that is we change
$$
\boldsymbol{X}^{T} \boldsymbol{X} \rightarrow \boldsymbol{X}^{T} \boldsymbol{X}+\lambda \boldsymbol{I},
$$
where \( \boldsymbol{I} \) is the identity matrix. When we discuss <b>Ridge</b> regression this is actually what we end up evaluating. The parameter \( \lambda \) is called a hyperparameter. More about this later.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec6">Basic math of the SVD </h2>
<p>
From standard linear algebra we know that a square matrix \( \boldsymbol{X} \) can be diagonalized if and only it is
a so-called <a href="https://en.wikipedia.org/wiki/Normal_matrix" target="_blank">normal matrix</a>, that is if \( \boldsymbol{X}\in {\mathbb{R}}^{n\times n} \)
we have \( \boldsymbol{X}\boldsymbol{X}^T=\boldsymbol{X}^T\boldsymbol{X} \) or if \( \boldsymbol{X}\in {\mathbb{C}}^{n\times n} \) we have \( \boldsymbol{X}\boldsymbol{X}^{\dagger}=\boldsymbol{X}^{\dagger}\boldsymbol{X} \).
The matrix has then a set of eigenpairs
$$
(\lambda_1,\boldsymbol{u}_1),\dots, (\lambda_n,\boldsymbol{u}_n),
$$
and the eigenvalues are given by the diagonal matrix
$$
\boldsymbol{\Sigma}=\mathrm{Diag}(\lambda_1, \dots,\lambda_n).
$$
The matrix \( \boldsymbol{X} \) can be written in terms of an orthogonal/unitary transformation \( \boldsymbol{U} \)
$$
\boldsymbol{X} = \boldsymbol{U}\boldsymbol{\Sigma}\boldsymbol{V}^T,
$$
with \( \boldsymbol{U}\boldsymbol{U}^T=\boldsymbol{I} \) or \( \boldsymbol{U}\boldsymbol{U}^{\dagger}=\boldsymbol{I} \).
<p>
Not all square matrices are diagonalizable. A matrix like the one discussed above
$$
\boldsymbol{X} = \begin{bmatrix}
1& -1 \\
1& -1\\
\end{bmatrix}
$$
is not diagonalizable, it is a so-called <a href="https://en.wikipedia.org/wiki/Defective_matrix" target="_blank">defective matrix</a>. It is easy to see that the condition
\( \boldsymbol{X}\boldsymbol{X}^T=\boldsymbol{X}^T\boldsymbol{X} \) is not fulfilled.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec7">The SVD, a Fantastic Algorithm </h2>
<p>
However, and this is the strength of the SVD algorithm, any general
matrix \( \boldsymbol{X} \) can be decomposed in terms of a diagonal matrix and
two orthogonal/unitary matrices. The <a href="https://en.wikipedia.org/wiki/Singular_value_decomposition" target="_blank">Singular Value Decompostion
(SVD) theorem</a>
states that a general \( m\times n \) matrix \( \boldsymbol{X} \) can be written in
terms of a diagonal matrix \( \boldsymbol{\Sigma} \) of dimensionality \( m\times n \)
and two orthognal matrices \( \boldsymbol{U} \) and \( \boldsymbol{V} \), where the first has
dimensionality \( m \times m \) and the last dimensionality \( n\times n \).
We have then
$$
\boldsymbol{X} = \boldsymbol{U}\boldsymbol{\Sigma}\boldsymbol{V}^T
$$
<p>
As an example, the above defective matrix can be decomposed as
$$
\boldsymbol{X} = \frac{1}{\sqrt{2}}\begin{bmatrix} 1& 1 \\ 1& -1\\ \end{bmatrix} \begin{bmatrix} 2& 0 \\ 0& 0\\ \end{bmatrix} \frac{1}{\sqrt{2}}\begin{bmatrix} 1& -1 \\ 1& 1\\ \end{bmatrix}=\boldsymbol{U}\boldsymbol{\Sigma}\boldsymbol{V}^T,
$$
<p>
with eigenvalues \( \sigma_1=2 \) and \( \sigma_2=0 \).
The SVD exits always!
<p>
The SVD
decomposition (singular values) gives eigenvalues
\( \sigma_i\geq\sigma_{i+1} \) for all \( i \) and for dimensions larger than \( i=p \), the
eigenvalues (singular values) are zero.
<p>
In the general case, where our design matrix \( \boldsymbol{X} \) has dimension
\( n\times p \), the matrix is thus decomposed into an \( n\times n \)
orthogonal matrix \( \boldsymbol{U} \), a \( p\times p \) orthogonal matrix \( \boldsymbol{V} \)
and a diagonal matrix \( \boldsymbol{\Sigma} \) with \( r=\mathrm{min}(n,p) \)
singular values \( \sigma_i\geq 0 \) on the main diagonal and zeros filling
the rest of the matrix. There are at most \( p \) singular values
assuming that \( n > p \). In our regression examples for the nuclear
masses and the equation of state this is indeed the case, while for
the Ising model we have \( p > n \). These are often cases that lead to
near singular or singular matrices.
<p>
The columns of \( \boldsymbol{U} \) are called the left singular vectors while the columns of \( \boldsymbol{V} \) are the right singular vectors.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec8">Economy-size SVD </h2>
<p>
If we assume that \( n > p \), then our matrix \( \boldsymbol{U} \) has dimension \( n
\times n \). The last \( n-p \) columns of \( \boldsymbol{U} \) become however
irrelevant in our calculations since they are multiplied with the
zeros in \( \boldsymbol{\Sigma} \).
<p>
The economy-size decomposition removes extra rows or columns of zeros
from the diagonal matrix of singular values, \( \boldsymbol{\Sigma} \), along with the columns
in either \( \boldsymbol{U} \) or \( \boldsymbol{V} \) that multiply those zeros in the expression.
Removing these zeros and columns can improve execution time
and reduce storage requirements without compromising the accuracy of
the decomposition.
<p>
If \( n > p \), we keep only the first \( p \) columns of \( \boldsymbol{U} \) and \( \boldsymbol{\Sigma} \) has dimension \( p\times p \).
If \( p > n \), then only the first \( n \) columns of \( \boldsymbol{V} \) are computed and \( \boldsymbol{\Sigma} \) has dimension \( n\times n \).
The \( n=p \) case is obvious, we retain the full SVD.
In general the economy-size SVD leads to less FLOPS and still conserving the desired accuracy.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec9">Codes for the SVD </h2>
<p>
<!-- code=python (!bc pycod) typeset with pygments style "default" -->
<div class="highlight" style="background: #f8f8f8"><pre style="line-height: 125%"><span></span><span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">numpy</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">np</span>
<span style="color: #408080; font-style: italic"># SVD inversion</span>
<span style="color: #008000; font-weight: bold">def</span> <span style="color: #0000FF">SVDinv</span>(A):
<span style="color: #BA2121; font-style: italic">&#39;&#39;&#39; Takes as input a numpy matrix A and returns inv(A) based on singular value decomposition (SVD).</span>
<span style="color: #BA2121; font-style: italic"> SVD is numerically more stable than the inversion algorithms provided by</span>
<span style="color: #BA2121; font-style: italic"> numpy and scipy.linalg at the cost of being slower.</span>
<span style="color: #BA2121; font-style: italic"> &#39;&#39;&#39;</span>
U, s, VT <span style="color: #666666">=</span> np<span style="color: #666666">.</span>linalg<span style="color: #666666">.</span>svd(A)
<span style="color: #408080; font-style: italic"># print(&#39;test U&#39;)</span>
<span style="color: #408080; font-style: italic"># print( (np.transpose(U) @ U - U @np.transpose(U)))</span>
<span style="color: #408080; font-style: italic"># print(&#39;test VT&#39;)</span>
<span style="color: #408080; font-style: italic"># print( (np.transpose(VT) @ VT - VT @np.transpose(VT)))</span>
<span style="color: #008000; font-weight: bold">print</span>(U)
<span style="color: #008000; font-weight: bold">print</span>(s)
<span style="color: #008000; font-weight: bold">print</span>(VT)
D <span style="color: #666666">=</span> np<span style="color: #666666">.</span>zeros((<span style="color: #008000">len</span>(U),<span style="color: #008000">len</span>(VT)))
<span style="color: #008000; font-weight: bold">for</span> i <span style="color: #AA22FF; font-weight: bold">in</span> <span style="color: #008000">range</span>(<span style="color: #666666">0</span>,<span style="color: #008000">len</span>(VT)):
D[i,i]<span style="color: #666666">=</span>s[i]
UT <span style="color: #666666">=</span> np<span style="color: #666666">.</span>transpose(U); V <span style="color: #666666">=</span> np<span style="color: #666666">.</span>transpose(VT); invD <span style="color: #666666">=</span> np<span style="color: #666666">.</span>linalg<span style="color: #666666">.</span>inv(D)
<span style="color: #008000; font-weight: bold">return</span> np<span style="color: #666666">.</span>matmul(V,np<span style="color: #666666">.</span>matmul(invD,UT))
X <span style="color: #666666">=</span> np<span style="color: #666666">.</span>array([ [<span style="color: #666666">1.0</span>, <span style="color: #666666">-1.0</span>, <span style="color: #666666">2.0</span>], [<span style="color: #666666">1.0</span>, <span style="color: #666666">0.0</span>, <span style="color: #666666">1.0</span>], [<span style="color: #666666">1.0</span>, <span style="color: #666666">2.0</span>, <span style="color: #666666">-1.0</span>], [<span style="color: #666666">1.0</span>, <span style="color: #666666">1.0</span>, <span style="color: #666666">0.0</span>] ])
<span style="color: #008000; font-weight: bold">print</span>(X)
A <span style="color: #666666">=</span> np<span style="color: #666666">.</span>transpose(X) @ X
<span style="color: #008000; font-weight: bold">print</span>(A)
<span style="color: #408080; font-style: italic"># Brute force inversion of super-collinear matrix</span>
<span style="color: #408080; font-style: italic">#B = np.linalg.inv(A)</span>
<span style="color: #408080; font-style: italic">#print(B)</span>
C <span style="color: #666666">=</span> SVDinv(A)
<span style="color: #008000; font-weight: bold">print</span>(C)
</pre></div>
<p>
The matrix \( \boldsymbol{X} \) has columns that are linearly dependent. The first
column is the row-wise sum of the other two columns. The rank of a
matrix (the column rank) is the dimension of space spanned by the
column vectors. The rank of the matrix is the number of linearly
independent columns, in this case just \( 2 \). We see this from the
singular values when running the above code. Running the standard
inversion algorithm for matrix inversion with \( \boldsymbol{X}^T\boldsymbol{X} \) results
in the program terminating due to a singular matrix.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec10">Mathematical Properties </h2>
<p>
There are several interesting mathematical properties which will be
relevant when we are going to discuss the differences between say
ordinary least squares (OLS) and <b>Ridge</b> regression.
<p>
We have from OLS that the parameters of the linear approximation are given by
$$
\boldsymbol{\tilde{y}} = \boldsymbol{X}\boldsymbol{\beta} = \boldsymbol{X}\left(\boldsymbol{X}^T\boldsymbol{X}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y}.
$$
<p>
The matrix to invert can be rewritten in terms of our SVD decomposition as
$$
\boldsymbol{X}^T\boldsymbol{X} = \boldsymbol{V}\boldsymbol{\Sigma}^T\boldsymbol{U}^T\boldsymbol{U}\boldsymbol{\Sigma}\boldsymbol{V}^T.
$$
Using the orthogonality properties of \( \boldsymbol{U} \) we have
$$
\boldsymbol{X}^T\boldsymbol{X} = \boldsymbol{V}\boldsymbol{\Sigma}^T\boldsymbol{\Sigma}\boldsymbol{V}^T = \boldsymbol{V}\boldsymbol{D}\boldsymbol{V}^T,
$$
with \( \boldsymbol{D} \) being a diagonal matrix with values along the diagonal given by the singular values squared.
<p>
This means that
$$
(\boldsymbol{X}^T\boldsymbol{X})\boldsymbol{V} = \boldsymbol{V}\boldsymbol{D},
$$
that is the eigenvectors of \( (\boldsymbol{X}^T\boldsymbol{X}) \) are given by the columns of the right singular matrix of \( \boldsymbol{X} \) and the eigenvalues are the squared singular values. It is easy to show (show this) that
$$
(\boldsymbol{X}\boldsymbol{X}^T)\boldsymbol{U} = \boldsymbol{U}\boldsymbol{D},
$$
that is, the eigenvectors of \( (\boldsymbol{X}\boldsymbol{X})^T \) are the columns of the left singular matrix and the eigenvalues are the same.
<p>
Going back to our OLS equation we have
$$
\boldsymbol{X}\boldsymbol{\beta} = \boldsymbol{X}\left(\boldsymbol{V}\boldsymbol{D}\boldsymbol{V}^T \right)^{-1}\boldsymbol{X}^T\boldsymbol{y}=\boldsymbol{U\Sigma V^T}\left(\boldsymbol{V}\boldsymbol{D}\boldsymbol{V}^T \right)^{-1}(\boldsymbol{U\Sigma V^T})^T\boldsymbol{y}=\boldsymbol{U}\boldsymbol{U}^T\boldsymbol{y}.
$$
We will come back to this expression when we discuss Ridge regression.
$$ \tilde{y}^{OLS}={\bf X}\hat{\beta}^{OLS}=\sum_{j=1}^p {\bf u}_j{\bf u}_j^T{\bf y}$$ and for Ridge we have&#160;
$$ \tilde{y}^{Ridge}={\bf X}\hat{\beta}^{Ridge}=\sum_{j=1}^p {\bf u}_j\frac{\sigma_j^2}{\sigma_j^2+\lambda}{\bf u}_j^T{\bf y}$$ .&#160;
<p>
It is indeed the economy-sized SVD, note the summation runs up tp $$p$$ only and not $$n$$.&#160;
<p>
Here we have that $${\bf X} = {\bf U}{\bf \Sigma}{\bf V}^T$$, with $$\Sigma$$ being an $$ n\times p$$ matrix and $${\bf V}$$ being a $$ p\times p$$ matrix. We also have assumed here that $$ n > p$$.&#160;
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec11">Friday September 12 </h2>
<p>
<a href="https://www.uio.no/studier/emner/matnat/fys/FYS-STK4155/h20/forelesningsvideoer/LectureSeptember11.mp4?vrtx=view-as-webpage" target="_blank">Video of Lecture</a> and <a href="https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/NotesSeptember11.pdf" target="_blank">handwritten notes</a>
<p>
More material will be added here, see handwritten notes also.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec12">Ridge and LASSO Regression </h2>
<p>
Let us remind ourselves about the expression for the standard Mean Squared Error (MSE) which we used to define our cost function and the equations for the ordinary least squares (OLS) method, that is
our optimization problem is
$$
{\displaystyle \min_{\boldsymbol{\beta}\in {\mathbb{R}}^{p}}}\frac{1}{n}\left\{\left(\boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta}\right)^T\left(\boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta}\right)\right\}.
$$
or we can state it as
$$
{\displaystyle \min_{\boldsymbol{\beta}\in
{\mathbb{R}}^{p}}}\frac{1}{n}\sum_{i=0}^{n-1}\left(y_i-\tilde{y}_i\right)^2=\frac{1}{n}\vert\vert \boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta}\vert\vert_2^2,
$$
where we have used the definition of a norm-2 vector, that is
$$
\vert\vert \boldsymbol{x}\vert\vert_2 = \sqrt{\sum_i x_i^2}.
$$
<p>
By minimizing the above equation with respect to the parameters
\( \boldsymbol{\beta} \) we could then obtain an analytical expression for the
parameters \( \boldsymbol{\beta} \). We can add a regularization parameter \( \lambda \) by
defining a new cost function to be optimized, that is
$$
{\displaystyle \min_{\boldsymbol{\beta}\in
{\mathbb{R}}^{p}}}\frac{1}{n}\vert\vert \boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta}\vert\vert_2^2+\lambda\vert\vert \boldsymbol{\beta}\vert\vert_2^2
$$
<p>
which leads to the Ridge regression minimization problem where we
require that \( \vert\vert \boldsymbol{\beta}\vert\vert_2^2\le t \), where \( t \) is
a finite number larger than zero. By defining
$$
C(\boldsymbol{X},\boldsymbol{\beta})=\frac{1}{n}\vert\vert \boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta}\vert\vert_2^2+\lambda\vert\vert \boldsymbol{\beta}\vert\vert_1,
$$
<p>
we have a new optimization equation
$$
{\displaystyle \min_{\boldsymbol{\beta}\in
{\mathbb{R}}^{p}}}\frac{1}{n}\vert\vert \boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta}\vert\vert_2^2+\lambda\vert\vert \boldsymbol{\beta}\vert\vert_1
$$
which leads to Lasso regression. Lasso stands for least absolute shrinkage and selection operator.
<p>
Here we have defined the norm-1 as
$$
\vert\vert \boldsymbol{x}\vert\vert_1 = \sum_i \vert x_i\vert.
$$
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec13">More on Ridge Regression </h2>
<p>
Using the matrix-vector expression for Ridge regression,
$$
C(\boldsymbol{X},\boldsymbol{\beta})=\frac{1}{n}\left\{(\boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta})^T(\boldsymbol{y}-\boldsymbol{X}\boldsymbol{\beta})\right\}+\lambda\boldsymbol{\beta}^T\boldsymbol{\beta},
$$
<p>
by taking the derivatives with respect to \( \boldsymbol{\beta} \) we obtain then
a slightly modified matrix inversion problem which for finite values
of \( \lambda \) does not suffer from singularity problems. We obtain
$$
\boldsymbol{\beta}^{\mathrm{Ridge}} = \left(\boldsymbol{X}^T\boldsymbol{X}+\lambda\boldsymbol{I}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y},
$$
<p>
with \( \boldsymbol{I} \) being a \( p\times p \) identity matrix with the constraint that
$$
\sum_{i=0}^{p-1} \beta_i^2 \leq t,
$$
<p>
with \( t \) a finite positive number.
<p>
We see that Ridge regression is nothing but the standard
OLS with a modified diagonal term added to \( \boldsymbol{X}^T\boldsymbol{X} \). The
consequences, in particular for our discussion of the bias-variance tradeoff
are rather interesting.
<p>
Furthermore, if we use the result above in terms of the SVD decomposition (our analysis was done for the OLS method), we had
$$
(\boldsymbol{X}\boldsymbol{X}^T)\boldsymbol{U} = \boldsymbol{U}\boldsymbol{D}.
$$
<p>
We can analyse the OLS solutions in terms of the eigenvectors (the columns) of the right singular value matrix \( \boldsymbol{U} \) as
$$
\boldsymbol{X}\boldsymbol{\beta} = \boldsymbol{X}\left(\boldsymbol{V}\boldsymbol{D}\boldsymbol{V}^T \right)^{-1}\boldsymbol{X}^T\boldsymbol{y}=\boldsymbol{U\Sigma V^T}\left(\boldsymbol{V}\boldsymbol{D}\boldsymbol{V}^T \right)^{-1}(\boldsymbol{U\Sigma V^T})^T\boldsymbol{y}=\boldsymbol{U}\boldsymbol{U}^T\boldsymbol{y}
$$
<p>
For Ridge regression this becomes
$$
\boldsymbol{X}\boldsymbol{\beta}^{\mathrm{Ridge}} = \boldsymbol{U\Sigma V^T}\left(\boldsymbol{V}\boldsymbol{D}\boldsymbol{V}^T+\lambda\boldsymbol{I} \right)^{-1}(\boldsymbol{U\Sigma V^T})^T\boldsymbol{y}=\sum_{j=0}^{p-1}\boldsymbol{u}_j\boldsymbol{u}_j^T\frac{\sigma_j^2}{\sigma_j^2+\lambda}\boldsymbol{y},
$$
<p>
with the vectors \( \boldsymbol{u}_j \) being the columns of \( \boldsymbol{U} \).
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec14">Interpreting the Ridge results </h2>
<p>
Since \( \lambda \geq 0 \), it means that compared to OLS, we have
$$
\frac{\sigma_j^2}{\sigma_j^2+\lambda} \leq 1.
$$
<p>
Ridge regression finds the coordinates of \( \boldsymbol{y} \) with respect to the
orthonormal basis \( \boldsymbol{U} \), it then shrinks the coordinates by
\( \frac{\sigma_j^2}{\sigma_j^2+\lambda} \). Recall that the SVD has
eigenvalues ordered in a descending way, that is \( \sigma_i \geq
\sigma_{i+1} \).
<p>
For small eigenvalues \( \sigma_i \) it means that their contributions become less important, a fact which can be used to reduce the number of degrees of freedom.
Actually, calculating the variance of \( \boldsymbol{X}\boldsymbol{v}_j \) shows that this quantity is equal to \( \sigma_j^2/n \).
With a parameter \( \lambda \) we can thus shrink the role of specific parameters.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec15">More interpretations </h2>
<p>
For the sake of simplicity, let us assume that the design matrix is orthonormal, that is
$$
\boldsymbol{X}^T\boldsymbol{X}=(\boldsymbol{X}^T\boldsymbol{X})^{-1} =\boldsymbol{I}.
$$
<p>
In this case the standard OLS results in
$$
\boldsymbol{\beta}^{\mathrm{OLS}} = \boldsymbol{X}^T\boldsymbol{y}=\sum_{i=0}^{p-1}\boldsymbol{u}_j\boldsymbol{u}_j^T\boldsymbol{y},
$$
<p>
and
$$
\boldsymbol{\beta}^{\mathrm{Ridge}} = \left(\boldsymbol{I}+\lambda\boldsymbol{I}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y}=\left(1+\lambda\right)^{-1}\boldsymbol{\beta}^{\mathrm{OLS}},
$$
<p>
that is the Ridge estimator scales the OLS estimator by the inverse of a factor \( 1+\lambda \), and
the Ridge estimator converges to zero when the hyperparameter goes to
infinity.
<p>
We will come back to more interpreations after we have gone through some of the statistical analysis part.
<p>
For more discussions of Ridge and Lasso regression, <a href="https://arxiv.org/abs/1509.09169" target="_blank">Wessel van Wieringen's</a> article is highly recommended.
Similarly, <a href="https://arxiv.org/abs/1803.08823" target="_blank">Mehta et al's article</a> is also recommended.
<p>
<!-- !split -->
<h2 id="___sec16">A better understanding of regularization </h2>
<p>
The parameter \( \lambda \) that we have introduced in the Ridge (and
Lasso as well) regression is often called a regularization parameter
or shrinkage parameter. It is common to call it a hyperparameter. What does it mean mathemtically?
<p>
Here we will first look at how to analyze the difference between the
standard OLS equations and the Ridge expressions in terms of a linear
algebra analysis using the SVD algorithm. Thereafter, we will link
(see the material on the bias-variance tradeoff below) these
observation to the statisical analysis of the results. In particular
we consider how the variance of the parameters \( \boldsymbol{\beta} \) is
affected by changing the parameter \( \lambda \).
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec17">Decomposing the OLS and Ridge expressions </h2>
<p>
We have our design matrix
\( \boldsymbol{X}\in {\mathbb{R}}^{n\times p} \). With the SVD we decompose it as
$$
\boldsymbol{X} = \boldsymbol{U\Sigma V^T},
$$
<p>
with \( \boldsymbol{U}\in {\mathbb{R}}^{n\times n} \), \( \boldsymbol{\Sigma}\in {\mathbb{R}}^{n\times p} \)
and \( \boldsymbol{V}\in {\mathbb{R}}^{p\times p} \).
<p>
The matrices \( \boldsymbol{U} \) and \( \boldsymbol{V} \) are unitary/orthonormal matrices, that is in case the matrices are real we have \( \boldsymbol{U}^T\boldsymbol{U}=\boldsymbol{U}\boldsymbol{U}^T=\boldsymbol{I} \) and \( \boldsymbol{V}^T\boldsymbol{V}=\boldsymbol{V}\boldsymbol{V}^T=\boldsymbol{I} \).
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec18">Introducing the Covariance and Correlation functions </h2>
<p>
Before we discuss the link between for example Ridge regression and the singular value decomposition, we need to remind ourselves about
the definition of the covariance and the correlation function. These are quantities
<p>
Suppose we have defined two vectors
\( \hat{x} \) and \( \hat{y} \) with \( n \) elements each. The covariance matrix \( \boldsymbol{C} \) is defined as
$$
\boldsymbol{C}[\boldsymbol{x},\boldsymbol{y}] = \begin{bmatrix} \mathrm{cov}[\boldsymbol{x},\boldsymbol{x}] & \mathrm{cov}[\boldsymbol{x},\boldsymbol{y}] \\
\mathrm{cov}[\boldsymbol{y},\boldsymbol{x}] & \mathrm{cov}[\boldsymbol{y},\boldsymbol{y}] \\
\end{bmatrix},
$$
where for example
$$
\mathrm{cov}[\boldsymbol{x},\boldsymbol{y}] =\frac{1}{n} \sum_{i=0}^{n-1}(x_i- \overline{x})(y_i- \overline{y}).
$$
With this definition and recalling that the variance is defined as
$$
\mathrm{var}[\boldsymbol{x}]=\frac{1}{n} \sum_{i=0}^{n-1}(x_i- \overline{x})^2,
$$
we can rewrite the covariance matrix as
$$
\boldsymbol{C}[\boldsymbol{x},\boldsymbol{y}] = \begin{bmatrix} \mathrm{var}[\boldsymbol{x}] & \mathrm{cov}[\boldsymbol{x},\boldsymbol{y}] \\
\mathrm{cov}[\boldsymbol{x},\boldsymbol{y}] & \mathrm{var}[\boldsymbol{y}] \\
\end{bmatrix}.
$$
<p>
The covariance takes values between zero and infinity and may thus
lead to problems with loss of numerical precision for particularly
large values. It is common to scale the covariance matrix by
introducing instead the correlation matrix defined via the so-called
correlation function
$$
\mathrm{corr}[\boldsymbol{x},\boldsymbol{y}]=\frac{\mathrm{cov}[\boldsymbol{x},\boldsymbol{y}]}{\sqrt{\mathrm{var}[\boldsymbol{x}] \mathrm{var}[\boldsymbol{y}]}}.
$$
<p>
The correlation function is then given by values \( \mathrm{corr}[\boldsymbol{x},\boldsymbol{y}]
\in [-1,1] \). This avoids eventual problems with too large values. We
can then define the correlation matrix for the two vectors \( \boldsymbol{x} \)
and \( \boldsymbol{y} \) as
$$
\boldsymbol{K}[\boldsymbol{x},\boldsymbol{y}] = \begin{bmatrix} 1 & \mathrm{corr}[\boldsymbol{x},\boldsymbol{y}] \\
\mathrm{corr}[\boldsymbol{y},\boldsymbol{x}] & 1 \\
\end{bmatrix},
$$
<p>
In the above example this is the function we constructed using <b>pandas</b>.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec19">Correlation Function and Design/Feature Matrix </h2>
<p>
In our derivation of the various regression algorithms like <b>Ordinary Least Squares</b> or <b>Ridge regression</b>
we defined the design/feature matrix \( \boldsymbol{X} \) as
$$
\boldsymbol{X}=\begin{bmatrix}
x_{0,0} & x_{0,1} & x_{0,2}& \dots & \dots x_{0,p-1}\\
x_{1,0} & x_{1,1} & x_{1,2}& \dots & \dots x_{1,p-1}\\
x_{2,0} & x_{2,1} & x_{2,2}& \dots & \dots x_{2,p-1}\\
\dots & \dots & \dots & \dots \dots & \dots \\
x_{n-2,0} & x_{n-2,1} & x_{n-2,2}& \dots & \dots x_{n-2,p-1}\\
x_{n-1,0} & x_{n-1,1} & x_{n-1,2}& \dots & \dots x_{n-1,p-1}\\
\end{bmatrix},
$$
with \( \boldsymbol{X}\in {\mathbb{R}}^{n\times p} \), with the predictors/features \( p \) refering to the column numbers and the
entries \( n \) being the row elements.
We can rewrite the design/feature matrix in terms of its column vectors as
$$
\boldsymbol{X}=\begin{bmatrix} \boldsymbol{x}_0 & \boldsymbol{x}_1 & \boldsymbol{x}_2 & \dots & \dots & \boldsymbol{x}_{p-1}\end{bmatrix},
$$
with a given vector
$$
\boldsymbol{x}_i^T = \begin{bmatrix}x_{0,i} & x_{1,i} & x_{2,i}& \dots & \dots x_{n-1,i}\end{bmatrix}.
$$
<p>
With these definitions, we can now rewrite our \( 2\times 2 \)
correaltion/covariance matrix in terms of a moe general design/feature
matrix \( \boldsymbol{X}\in {\mathbb{R}}^{n\times p} \). This leads to a \( p\times p \)
covariance matrix for the vectors \( \boldsymbol{x}_i \) with \( i=0,1,\dots,p-1 \)
$$
\boldsymbol{C}[\boldsymbol{x}] = \begin{bmatrix}
\mathrm{var}[\boldsymbol{x}_0] & \mathrm{cov}[\boldsymbol{x}_0,\boldsymbol{x}_1] & \mathrm{cov}[\boldsymbol{x}_0,\boldsymbol{x}_2] & \dots & \dots & \mathrm{cov}[\boldsymbol{x}_0,\boldsymbol{x}_{p-1}]\\
\mathrm{cov}[\boldsymbol{x}_1,\boldsymbol{x}_0] & \mathrm{var}[\boldsymbol{x}_1] & \mathrm{cov}[\boldsymbol{x}_1,\boldsymbol{x}_2] & \dots & \dots & \mathrm{cov}[\boldsymbol{x}_1,\boldsymbol{x}_{p-1}]\\
\mathrm{cov}[\boldsymbol{x}_2,\boldsymbol{x}_0] & \mathrm{cov}[\boldsymbol{x}_2,\boldsymbol{x}_1] & \mathrm{var}[\boldsymbol{x}_2] & \dots & \dots & \mathrm{cov}[\boldsymbol{x}_2,\boldsymbol{x}_{p-1}]\\
\dots & \dots & \dots & \dots & \dots & \dots \\
\dots & \dots & \dots & \dots & \dots & \dots \\
\mathrm{cov}[\boldsymbol{x}_{p-1},\boldsymbol{x}_0] & \mathrm{cov}[\boldsymbol{x}_{p-1},\boldsymbol{x}_1] & \mathrm{cov}[\boldsymbol{x}_{p-1},\boldsymbol{x}_{2}] & \dots & \dots & \mathrm{var}[\boldsymbol{x}_{p-1}]\\
\end{bmatrix},
$$
and the correlation matrix
$$
\boldsymbol{K}[\boldsymbol{x}] = \begin{bmatrix}
1 & \mathrm{corr}[\boldsymbol{x}_0,\boldsymbol{x}_1] & \mathrm{corr}[\boldsymbol{x}_0,\boldsymbol{x}_2] & \dots & \dots & \mathrm{corr}[\boldsymbol{x}_0,\boldsymbol{x}_{p-1}]\\
\mathrm{corr}[\boldsymbol{x}_1,\boldsymbol{x}_0] & 1 & \mathrm{corr}[\boldsymbol{x}_1,\boldsymbol{x}_2] & \dots & \dots & \mathrm{corr}[\boldsymbol{x}_1,\boldsymbol{x}_{p-1}]\\
\mathrm{corr}[\boldsymbol{x}_2,\boldsymbol{x}_0] & \mathrm{corr}[\boldsymbol{x}_2,\boldsymbol{x}_1] & 1 & \dots & \dots & \mathrm{corr}[\boldsymbol{x}_2,\boldsymbol{x}_{p-1}]\\
\dots & \dots & \dots & \dots & \dots & \dots \\
\dots & \dots & \dots & \dots & \dots & \dots \\
\mathrm{corr}[\boldsymbol{x}_{p-1},\boldsymbol{x}_0] & \mathrm{corr}[\boldsymbol{x}_{p-1},\boldsymbol{x}_1] & \mathrm{corr}[\boldsymbol{x}_{p-1},\boldsymbol{x}_{2}] & \dots & \dots & 1\\
\end{bmatrix},
$$
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec20">Covariance Matrix Examples </h2>
<p>
The Numpy function <b>np.cov</b> calculates the covariance elements using
the factor \( 1/(n-1) \) instead of \( 1/n \) since it assumes we do not have
the exact mean values. The following simple function uses the
<b>np.vstack</b> function which takes each vector of dimension \( 1\times n \)
and produces a \( 2\times n \) matrix \( \boldsymbol{W} \)
$$
\boldsymbol{W} = \begin{bmatrix} x_0 & y_0 \\
x_1 & y_1 \\
x_2 & y_2\\
\dots & \dots \\
x_{n-2} & y_{n-2}\\
x_{n-1} & y_{n-1} &
\end{bmatrix},
$$
<p>
which in turn is converted into into the \( 2\times 2 \) covariance matrix
\( \boldsymbol{C} \) via the Numpy function <b>np.cov()</b>. We note that we can also calculate
the mean value of each set of samples \( \boldsymbol{x} \) etc using the Numpy
function <b>np.mean(x)</b>. We can also extract the eigenvalues of the
covariance matrix through the <b>np.linalg.eig()</b> function.
<p>
<!-- code=python (!bc pycod) typeset with pygments style "default" -->
<div class="highlight" style="background: #f8f8f8"><pre style="line-height: 125%"><span></span><span style="color: #408080; font-style: italic"># Importing various packages</span>
<span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">numpy</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">np</span>
n <span style="color: #666666">=</span> <span style="color: #666666">100</span>
x <span style="color: #666666">=</span> np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>normal(size<span style="color: #666666">=</span>n)
<span style="color: #008000; font-weight: bold">print</span>(np<span style="color: #666666">.</span>mean(x))
y <span style="color: #666666">=</span> <span style="color: #666666">4+3*</span>x<span style="color: #666666">+</span>np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>normal(size<span style="color: #666666">=</span>n)
<span style="color: #008000; font-weight: bold">print</span>(np<span style="color: #666666">.</span>mean(y))
W <span style="color: #666666">=</span> np<span style="color: #666666">.</span>vstack((x, y))
C <span style="color: #666666">=</span> np<span style="color: #666666">.</span>cov(W)
<span style="color: #008000; font-weight: bold">print</span>(C)
</pre></div>
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec21">Correlation Matrix </h2>
<p>
The previous example can be converted into the correlation matrix by
simply scaling the matrix elements with the variances. We should also
subtract the mean values for each column. This leads to the following
code which sets up the correlations matrix for the previous example in
a more brute force way. Here we scale the mean values for each column of the design matrix, calculate the relevant mean values and variances and then finally set up the \( 2\times 2 \) correlation matrix (since we have only two vectors).
<p>
<!-- code=python (!bc pycod) typeset with pygments style "default" -->
<div class="highlight" style="background: #f8f8f8"><pre style="line-height: 125%"><span></span><span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">numpy</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">np</span>
n <span style="color: #666666">=</span> <span style="color: #666666">100</span>
<span style="color: #408080; font-style: italic"># define two vectors </span>
x <span style="color: #666666">=</span> np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>random(size<span style="color: #666666">=</span>n)
y <span style="color: #666666">=</span> <span style="color: #666666">4+3*</span>x<span style="color: #666666">+</span>np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>normal(size<span style="color: #666666">=</span>n)
<span style="color: #408080; font-style: italic">#scaling the x and y vectors </span>
x <span style="color: #666666">=</span> x <span style="color: #666666">-</span> np<span style="color: #666666">.</span>mean(x)
y <span style="color: #666666">=</span> y <span style="color: #666666">-</span> np<span style="color: #666666">.</span>mean(y)
variance_x <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sum(x<span style="color: #AA22FF">@x</span>)<span style="color: #666666">/</span>n
variance_y <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sum(y<span style="color: #AA22FF">@y</span>)<span style="color: #666666">/</span>n
<span style="color: #008000; font-weight: bold">print</span>(variance_x)
<span style="color: #008000; font-weight: bold">print</span>(variance_y)
cov_xy <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sum(x<span style="color: #AA22FF">@y</span>)<span style="color: #666666">/</span>n
cov_xx <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sum(x<span style="color: #AA22FF">@x</span>)<span style="color: #666666">/</span>n
cov_yy <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sum(y<span style="color: #AA22FF">@y</span>)<span style="color: #666666">/</span>n
C <span style="color: #666666">=</span> np<span style="color: #666666">.</span>zeros((<span style="color: #666666">2</span>,<span style="color: #666666">2</span>))
C[<span style="color: #666666">0</span>,<span style="color: #666666">0</span>]<span style="color: #666666">=</span> cov_xx<span style="color: #666666">/</span>variance_x
C[<span style="color: #666666">1</span>,<span style="color: #666666">1</span>]<span style="color: #666666">=</span> cov_yy<span style="color: #666666">/</span>variance_y
C[<span style="color: #666666">0</span>,<span style="color: #666666">1</span>]<span style="color: #666666">=</span> cov_xy<span style="color: #666666">/</span>np<span style="color: #666666">.</span>sqrt(variance_y<span style="color: #666666">*</span>variance_x)
C[<span style="color: #666666">1</span>,<span style="color: #666666">0</span>]<span style="color: #666666">=</span> C[<span style="color: #666666">0</span>,<span style="color: #666666">1</span>]
<span style="color: #008000; font-weight: bold">print</span>(C)
</pre></div>
<p>
We see that the matrix elements along the diagonal are one as they
should be and that the matrix is symmetric. Furthermore, diagonalizing
this matrix we easily see that it is a positive definite matrix.
<p>
The above procedure with <b>numpy</b> can be made more compact if we use <b>pandas</b>.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec22">Correlation Matrix with Pandas </h2>
<p>
We whow here how we can set up the correlation matrix using <b>pandas</b>, as done in this simple code
<p>
<!-- code=python (!bc pycod) typeset with pygments style "default" -->
<div class="highlight" style="background: #f8f8f8"><pre style="line-height: 125%"><span></span><span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">numpy</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">np</span>
<span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">pandas</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">pd</span>
n <span style="color: #666666">=</span> <span style="color: #666666">10</span>
x <span style="color: #666666">=</span> np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>normal(size<span style="color: #666666">=</span>n)
x <span style="color: #666666">=</span> x <span style="color: #666666">-</span> np<span style="color: #666666">.</span>mean(x)
y <span style="color: #666666">=</span> <span style="color: #666666">4+3*</span>x<span style="color: #666666">+</span>np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>normal(size<span style="color: #666666">=</span>n)
y <span style="color: #666666">=</span> y <span style="color: #666666">-</span> np<span style="color: #666666">.</span>mean(y)
X <span style="color: #666666">=</span> (np<span style="color: #666666">.</span>vstack((x, y)))<span style="color: #666666">.</span>T
<span style="color: #008000; font-weight: bold">print</span>(X)
Xpd <span style="color: #666666">=</span> pd<span style="color: #666666">.</span>DataFrame(X)
<span style="color: #008000; font-weight: bold">print</span>(Xpd)
correlation_matrix <span style="color: #666666">=</span> Xpd<span style="color: #666666">.</span>corr()
<span style="color: #008000; font-weight: bold">print</span>(correlation_matrix)
</pre></div>
<p>
We expand this model to the Franke function discussed above.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec23">Correlation Matrix with Pandas and the Franke function </h2>
<p>
<!-- code=python (!bc pycod) typeset with pygments style "default" -->
<div class="highlight" style="background: #f8f8f8"><pre style="line-height: 125%"><span></span><span style="color: #408080; font-style: italic"># Common imports</span>
<span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">numpy</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">np</span>
<span style="color: #008000; font-weight: bold">import</span> <span style="color: #0000FF; font-weight: bold">pandas</span> <span style="color: #008000; font-weight: bold">as</span> <span style="color: #0000FF; font-weight: bold">pd</span>
<span style="color: #008000; font-weight: bold">def</span> <span style="color: #0000FF">FrankeFunction</span>(x,y):
term1 <span style="color: #666666">=</span> <span style="color: #666666">0.75*</span>np<span style="color: #666666">.</span>exp(<span style="color: #666666">-</span>(<span style="color: #666666">0.25*</span>(<span style="color: #666666">9*</span>x<span style="color: #666666">-2</span>)<span style="color: #666666">**2</span>) <span style="color: #666666">-</span> <span style="color: #666666">0.25*</span>((<span style="color: #666666">9*</span>y<span style="color: #666666">-2</span>)<span style="color: #666666">**2</span>))
term2 <span style="color: #666666">=</span> <span style="color: #666666">0.75*</span>np<span style="color: #666666">.</span>exp(<span style="color: #666666">-</span>((<span style="color: #666666">9*</span>x<span style="color: #666666">+1</span>)<span style="color: #666666">**2</span>)<span style="color: #666666">/49.0</span> <span style="color: #666666">-</span> <span style="color: #666666">0.1*</span>(<span style="color: #666666">9*</span>y<span style="color: #666666">+1</span>))
term3 <span style="color: #666666">=</span> <span style="color: #666666">0.5*</span>np<span style="color: #666666">.</span>exp(<span style="color: #666666">-</span>(<span style="color: #666666">9*</span>x<span style="color: #666666">-7</span>)<span style="color: #666666">**2/4.0</span> <span style="color: #666666">-</span> <span style="color: #666666">0.25*</span>((<span style="color: #666666">9*</span>y<span style="color: #666666">-3</span>)<span style="color: #666666">**2</span>))
term4 <span style="color: #666666">=</span> <span style="color: #666666">-0.2*</span>np<span style="color: #666666">.</span>exp(<span style="color: #666666">-</span>(<span style="color: #666666">9*</span>x<span style="color: #666666">-4</span>)<span style="color: #666666">**2</span> <span style="color: #666666">-</span> (<span style="color: #666666">9*</span>y<span style="color: #666666">-7</span>)<span style="color: #666666">**2</span>)
<span style="color: #008000; font-weight: bold">return</span> term1 <span style="color: #666666">+</span> term2 <span style="color: #666666">+</span> term3 <span style="color: #666666">+</span> term4
<span style="color: #008000; font-weight: bold">def</span> <span style="color: #0000FF">create_X</span>(x, y, n ):
<span style="color: #008000; font-weight: bold">if</span> <span style="color: #008000">len</span>(x<span style="color: #666666">.</span>shape) <span style="color: #666666">&gt;</span> <span style="color: #666666">1</span>:
x <span style="color: #666666">=</span> np<span style="color: #666666">.</span>ravel(x)
y <span style="color: #666666">=</span> np<span style="color: #666666">.</span>ravel(y)
N <span style="color: #666666">=</span> <span style="color: #008000">len</span>(x)
l <span style="color: #666666">=</span> <span style="color: #008000">int</span>((n<span style="color: #666666">+1</span>)<span style="color: #666666">*</span>(n<span style="color: #666666">+2</span>)<span style="color: #666666">/2</span>) <span style="color: #408080; font-style: italic"># Number of elements in beta</span>
X <span style="color: #666666">=</span> np<span style="color: #666666">.</span>ones((N,l))
<span style="color: #008000; font-weight: bold">for</span> i <span style="color: #AA22FF; font-weight: bold">in</span> <span style="color: #008000">range</span>(<span style="color: #666666">1</span>,n<span style="color: #666666">+1</span>):
q <span style="color: #666666">=</span> <span style="color: #008000">int</span>((i)<span style="color: #666666">*</span>(i<span style="color: #666666">+1</span>)<span style="color: #666666">/2</span>)
<span style="color: #008000; font-weight: bold">for</span> k <span style="color: #AA22FF; font-weight: bold">in</span> <span style="color: #008000">range</span>(i<span style="color: #666666">+1</span>):
X[:,q<span style="color: #666666">+</span>k] <span style="color: #666666">=</span> (x<span style="color: #666666">**</span>(i<span style="color: #666666">-</span>k))<span style="color: #666666">*</span>(y<span style="color: #666666">**</span>k)
<span style="color: #008000; font-weight: bold">return</span> X
<span style="color: #408080; font-style: italic"># Making meshgrid of datapoints and compute Franke&#39;s function</span>
n <span style="color: #666666">=</span> <span style="color: #666666">4</span>
N <span style="color: #666666">=</span> <span style="color: #666666">100</span>
x <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sort(np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>uniform(<span style="color: #666666">0</span>, <span style="color: #666666">1</span>, N))
y <span style="color: #666666">=</span> np<span style="color: #666666">.</span>sort(np<span style="color: #666666">.</span>random<span style="color: #666666">.</span>uniform(<span style="color: #666666">0</span>, <span style="color: #666666">1</span>, N))
z <span style="color: #666666">=</span> FrankeFunction(x, y)
X <span style="color: #666666">=</span> create_X(x, y, n<span style="color: #666666">=</span>n)
Xpd <span style="color: #666666">=</span> pd<span style="color: #666666">.</span>DataFrame(X)
<span style="color: #408080; font-style: italic"># subtract the mean values and set up the covariance matrix</span>
Xpd <span style="color: #666666">=</span> Xpd <span style="color: #666666">-</span> Xpd<span style="color: #666666">.</span>mean()
covariance_matrix <span style="color: #666666">=</span> Xpd<span style="color: #666666">.</span>cov()
<span style="color: #008000; font-weight: bold">print</span>(covariance_matrix)
</pre></div>
<p>
We note here that the covariance is zero for the first rows and
columns since all matrix elements in the design matrix were set to one
(we are fitting the function in terms of a polynomial of degree \( n \)).
<p>
This means that the variance for these elements will be zero and will
cause problems when we set up the correlation matrix. We can simply
drop these elements and construct a correlation
matrix without these elements.
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec24">Rewriting the Covariance and/or Correlation Matrix </h2>
<p>
We can rewrite the covariance matrix in a more compact form in terms of the design/feature matrix \( \boldsymbol{X} \) as
$$
\boldsymbol{C}[\boldsymbol{x}] = \frac{1}{n}\boldsymbol{X}^T\boldsymbol{X}= \mathbb{E}[\boldsymbol{X}^T\boldsymbol{X}].
$$
<p>
To see this let us simply look at a design matrix \( \boldsymbol{X}\in {\mathbb{R}}^{2\times 2} \)
$$
\boldsymbol{X}=\begin{bmatrix}
x_{00} & x_{01}\\
x_{10} & x_{11}\\
\end{bmatrix}=\begin{bmatrix}
\boldsymbol{x}_{0} & \boldsymbol{x}_{1}\\
\end{bmatrix}.
$$
<p>
If we then compute the expectation value
$$
\mathbb{E}[\boldsymbol{X}^T\boldsymbol{X}] = \frac{1}{n}\boldsymbol{X}^T\boldsymbol{X}=\begin{bmatrix}
x_{00}^2+x_{01}^2 & x_{00}x_{10}+x_{01}x_{11}\\
x_{10}x_{00}+x_{11}x_{01} & x_{10}^2+x_{11}^2\\
\end{bmatrix},
$$
which is just
$$
\boldsymbol{C}[\boldsymbol{x}_0,\boldsymbol{x}_1] = \boldsymbol{C}[\boldsymbol{x}]=\begin{bmatrix} \mathrm{var}[\boldsymbol{x}_0] & \mathrm{cov}[\boldsymbol{x}_0,\boldsymbol{x}_1] \\
\mathrm{cov}[\boldsymbol{x}_1,\boldsymbol{x}_0] & \mathrm{var}[\boldsymbol{x}_1] \\
\end{bmatrix},
$$
where we wrote $$\boldsymbol{C}[\boldsymbol{x}_0,\boldsymbol{x}_1] = \boldsymbol{C}[\boldsymbol{x}]$$ to indicate that this the covariance of the vectors \( \boldsymbol{x} \) of the design/feature matrix \( \boldsymbol{X} \).
<p>
It is easy to generalize this to a matrix \( \boldsymbol{X}\in {\mathbb{R}}^{n\times p} \).
<p>
<!-- !split --><br><br><br><br><br><br><br><br><br><br>
<h2 id="___sec25">Linking with SVD </h2>
<!-- ------------------- end of main content --------------- -->
<center style="font-size:80%">
<!-- copyright --> &copy; 1999-2020, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license
</center>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink