Chapter 4

The QR Algorithm

The QR algorithm computes a Schur decomposition of a matrix. It is certainly one of the

most important algorithm in eigenvalue computations [9]. However, it is applied to dense
(or: full) matrices only.
The QR algorithm consists of two separate stages. First, by means of a similarity
transformation, the original matrix is transformed in a finite number of steps to Hessenberg
form or – in the Hermitian/symmetric case – to real tridiagonal form. This first stage of
the algorithm prepares its second stage, the actual QR iterations that are applied to the
Hessenberg or tridiagonal matrix. The overall complexity (number of floating points) of
the algorithm is O(n3 ), which we will see is not entirely trivial to obtain.
The major limitation of the QR algorithm is that already the first stage generates
usually complete fill-in in general sparse matrices. It can therefore not be applied to large
sparse matrices, simply because of excessive memory requirements. On the other hand,
the QR algorithm computes all eigenvalues (and eventually eigenvectors) which is rarely
desired in sparse matrix computations anyway.
The treatment of the QR algorithm in these lecture notes on large scale eigenvalue
computation is justified in two respects. First, there are of course large or even huge dense
eigenvalue problems. Second, the QR algorithm is employed in most other algorithms to
solve ‘internal’ small auxiliary eigenvalue problems.

4.1 The basic QR algorithm

In 1958 Rutishauser [10] of ETH Zurich experimented with a similar algorithm that we are
going to present, but based on the LR factorization, i.e., based on Gaussian elimination
without pivoting. That algorithm was not successful as the LR factorization (nowadays
called LU factorization) is not stable without pivoting. Francis [5] noticed that the QR
factorization would be the preferred choice and devised the QR algorithm with many of
the bells and whistles used nowadays.
Before presenting the complete picture, we start with a basic iteration, given in Algo-
rithm 4.1, discuss its properties and improve on it step by step until we arrive at Francis’
algorithm.
We notice first that

(4.1) Ak = Rk Qk = Q∗k Ak−1 Qk ,

and hence Ak and Ak−1 are unitarily similar. The matrix sequence {Ak } converges (under
certain assumptions) towards an upper triangular matrix [11]. Let us assume that the

63
64 CHAPTER 4. THE QR ALGORITHM
Algorithm 4.1 Basic QR algorithm
1: Let A ∈ Cn×n . This algorithm computes an upper triangular matrix T and a unitary
matrix U such that A = U T U ∗ is the Schur decomposition of A.
2: Set A0 := A and U0 = I.
3: for k = 1, 2, . . . do
4: Ak−1 =: Qk Rk ; /* QR factorization */
5: Ak := Rk Qk ;
6: Uk := Uk−1 Qk ; /* Update transformation matrix */
7: end for
8: Set T := A∞ and U := U∞ .

eigenvalues are mutually different in magnitude and we can therefore number the eigen-
values such that |λ1 | > |λ2 | > · · · > |λn |. Then – as we will show in Chapter 8 – the
elements of Ak below the diagonal converge to zero like
(k)
(4.2) |aij | = O(|λi /λj |k ), i > j.

From (4.1) we see that

(4.3) Ak = Q∗k Ak−1 Qk = Q∗k Q∗k−1 Ak−2 Qk−1 Qk = · · · = Q∗k · · · Q∗1 A0 Q1 · · · Qk .

| {z }
Uk

With the same assumption on the eigenvalues, Ak tends to an upper triangular matrix
and Uk converges to the matrix of Schur vectors.

4.1.1 Numerical experiments

We conduct two Matlab experiments to illustrate the convergence rate given in (4.2). To
that end, we construct a random 4 × 4 matrix with eigenvalues 1, 2, 3, and 4.

D = diag([4 3 2 1]);
rand(’seed’,0);
format short e
S=rand(4); S = (S - .5)*2;
A = S*D/S % A_0 = A = S*D*S^{-1}
for i=1:20,
[Q,R] = qr(A); A = R*Q
end

This yields the matrix sequence

A( 0) = [ -4.4529e-01 4.9063e+00 -8.7871e-01 6.3036e+00]

[ -6.3941e+00 1.3354e+01 1.6668e+00 1.1945e+01]
[ 3.6842e+00 -6.6617e+00 -6.0021e-02 -7.0043e+00]
[ 3.1209e+00 -5.2052e+00 -1.4130e+00 -2.8484e+00]

A( 1) = [ 5.9284e+00 1.6107e+00 9.3153e-01 -2.2056e+01]

[ -1.5294e+00 1.8630e+00 2.0428e+00 6.5900e+00]
[ 1.9850e-01 2.5660e-01 1.7088e+00 1.2184e+00]
[ 2.4815e-01 1.5265e-01 2.6924e-01 4.9975e-01]

A( 2) = [ 4.7396e+00 1.4907e+00 -2.1236e+00 2.3126e+01]

4.1. THE BASIC QR ALGORITHM 65
[ -4.3101e-01 2.4307e+00 2.2544e+00 -8.2867e-01]
[ 1.2803e-01 2.4287e-01 1.6398e+00 -1.8290e+00]
[ -4.8467e-02 -5.8164e-02 -1.0994e-01 1.1899e+00]

A( 3) = [ 4.3289e+00 1.0890e+00 -3.9478e+00 -2.2903e+01]

[ -1.8396e-01 2.7053e+00 1.9060e+00 -1.2062e+00]
[ 6.7951e-02 1.7100e-01 1.6852e+00 2.5267e+00]
[ 1.3063e-02 2.2630e-02 7.9186e-02 1.2805e+00]

A( 4) = [ 4.1561e+00 7.6418e-01 -5.1996e+00 2.2582e+01]

[ -9.4175e-02 2.8361e+00 1.5788e+00 2.0983e+00]
[ 3.5094e-02 1.1515e-01 1.7894e+00 -2.9819e+00]
[ -3.6770e-03 -8.7212e-03 -5.7793e-02 1.2184e+00]

A( 5) = [ 4.0763e+00 5.2922e-01 -6.0126e+00 -2.2323e+01]

[ -5.3950e-02 2.9035e+00 1.3379e+00 -2.5358e+00]
[ 1.7929e-02 7.7393e-02 1.8830e+00 3.2484e+00]
[ 1.0063e-03 3.2290e-03 3.7175e-02 1.1372e+00]

A( 6) = [ 4.0378e+00 3.6496e-01 -6.4924e+00 2.2149e+01]

[ -3.3454e-02 2.9408e+00 1.1769e+00 2.7694e+00]
[ 9.1029e-03 5.2173e-02 1.9441e+00 -3.4025e+00]
[ -2.6599e-04 -1.1503e-03 -2.1396e-02 1.0773e+00]

A( 7) = [ 4.0189e+00 2.5201e-01 -6.7556e+00 -2.2045e+01]

[ -2.1974e-02 2.9627e+00 1.0736e+00 -2.9048e+00]
[ 4.6025e-03 3.5200e-02 1.9773e+00 3.4935e+00]
[ 6.8584e-05 3.9885e-04 1.1481e-02 1.0411e+00]

A( 8) = [ 4.0095e+00 1.7516e-01 -6.8941e+00 2.1985e+01]

[ -1.5044e-02 2.9761e+00 1.0076e+00 2.9898e+00]
[ 2.3199e-03 2.3720e-02 1.9932e+00 -3.5486e+00]
[ -1.7427e-05 -1.3602e-04 -5.9304e-03 1.0212e+00]

A( 9) = [ 4.0048e+00 1.2329e-01 -6.9655e+00 -2.1951e+01]

[ -1.0606e-02 2.9845e+00 9.6487e-01 -3.0469e+00]
[ 1.1666e-03 1.5951e-02 1.9999e+00 3.5827e+00]
[ 4.3933e-06 4.5944e-05 3.0054e-03 1.0108e+00]

A(10) = [ 4.0024e+00 8.8499e-02 -7.0021e+00 2.1931e+01]

[ -7.6291e-03 2.9899e+00 9.3652e-01 3.0873e+00]
[ 5.8564e-04 1.0704e-02 2.0023e+00 -3.6041e+00]
[ -1.1030e-06 -1.5433e-05 -1.5097e-03 1.0054e+00]

A(11) = [ 4.0013e+00 6.5271e-02 -7.0210e+00 -2.1920e+01]

[ -5.5640e-03 2.9933e+00 9.1729e-01 -3.1169e+00]
[ 2.9364e-04 7.1703e-03 2.0027e+00 3.6177e+00]
[ 2.7633e-07 5.1681e-06 7.5547e-04 1.0027e+00]

A(12) = [ 4.0007e+00 4.9824e-02 -7.0308e+00 2.1912e+01]

[ -4.0958e-03 2.9956e+00 9.0396e-01 3.1390e+00]
[ 1.4710e-04 4.7964e-03 2.0024e+00 -3.6265e+00]
[ -6.9154e-08 -1.7274e-06 -3.7751e-04 1.0014e+00]

A(13) = [ 4.0003e+00 3.9586e-02 -7.0360e+00 -2.1908e+01]

[ -3.0339e-03 2.9971e+00 8.9458e-01 -3.1558e+00]
[ 7.3645e-05 3.2052e-03 2.0019e+00 3.6322e+00]
[ 1.7298e-08 5.7677e-07 1.8857e-04 1.0007e+00]

A(14) = [ 4.0002e+00 3.2819e-02 -7.0388e+00 2.1905e+01]

66 CHAPTER 4. THE QR ALGORITHM
[ -2.2566e-03 2.9981e+00 8.8788e-01 3.1686e+00]
[ 3.6855e-05 2.1402e-03 2.0014e+00 -3.6359e+00]
[ -4.3255e-09 -1.9245e-07 -9.4197e-05 1.0003e+00]

A(15) = [ 4.0001e+00 2.8358e-02 -7.0404e+00 -2.1902e+01]

[ -1.6832e-03 2.9987e+00 8.8305e-01 -3.1784e+00]
[ 1.8438e-05 1.4284e-03 2.0010e+00 3.6383e+00]
[ 1.0815e-09 6.4192e-08 4.7062e-05 1.0002e+00]

A(16) = [ 4.0001e+00 2.5426e-02 -7.0413e+00 2.1901e+01]

[ -1.2577e-03 2.9991e+00 8.7953e-01 3.1859e+00]
[ 9.2228e-06 9.5295e-04 2.0007e+00 -3.6399e+00]
[ -2.7039e-10 -2.1406e-08 -2.3517e-05 1.0001e+00]

A(17) = [ 4.0000e+00 2.3503e-02 -7.0418e+00 -2.1900e+01]

[ -9.4099e-04 2.9994e+00 8.7697e-01 -3.1917e+00]
[ 4.6126e-06 6.3562e-04 2.0005e+00 3.6409e+00]
[ 6.7600e-11 7.1371e-09 1.1754e-05 1.0000e+00]

A(18) = [ 4.0000e+00 2.2246e-02 -7.0422e+00 2.1899e+01]

[ -7.0459e-04 2.9996e+00 8.7508e-01 3.1960e+00]
[ 2.3067e-06 4.2388e-04 2.0003e+00 -3.6416e+00]
[ -1.6900e-11 -2.3794e-09 -5.8750e-06 1.0000e+00]

A(19) = [ 4.0000e+00 2.1427e-02 -7.0424e+00 -2.1898e+01]

[ -5.2787e-04 2.9997e+00 8.7369e-01 -3.1994e+00]
[ 1.1535e-06 2.8265e-04 2.0002e+00 3.6421e+00]
[ 4.2251e-12 7.9321e-10 2.9369e-06 1.0000e+00]

A(20) = [ 4.0000e+00 2.0896e-02 -7.0425e+00 2.1898e+01]

[ -3.9562e-04 2.9998e+00 8.7266e-01 3.2019e+00]
[ 5.7679e-07 1.8846e-04 2.0002e+00 -3.6424e+00]
[ -1.0563e-12 -2.6442e-10 -1.4682e-06 1.0000e+00]

Looking at the element-wise quotients of the last two matrices one recognizes the conver-
gence rates claimed in (4.2).
A(20)./A(19) = [ 1.0000 0.9752 1.0000 -1.0000]
[ 0.7495 1.0000 0.9988 -1.0008]
[ 0.5000 0.6668 1.0000 -1.0001]
[ -0.2500 -0.3334 -0.4999 1.0000]

The elements above and on the diagonal are relatively stable.

If we run the same little Matlab script but with the initial diagonal matrix D replaced
by
D = diag([5 2 2 1]);
then we obtain
A(19) = [ 5.0000e+00 4.0172e+00 -9.7427e+00 -3.3483e+01]
[ -4.2800e-08 2.0000e+00 2.1100e-05 -4.3247e+00]
[ 1.3027e-08 7.0605e-08 2.0000e+00 2.1769e+00]
[ 8.0101e-14 -2.4420e-08 4.8467e-06 1.0000e+00]

A(20) = [ 5.0000e+00 4.0172e+00 -9.7427e+00 3.3483e+01]

[ -1.7120e-08 2.0000e+00 1.0536e-05 4.3247e+00]
[ 5.2106e-09 3.3558e-08 2.0000e+00 -2.1769e+00]
[ -1.6020e-14 1.2210e-08 -2.4234e-06 1.0000e+00]

So, again the eigenvalues are visible on the diagonal of A20 . The element-wise quotients
of A20 relative to A19 are
4.2. THE HESSENBERG QR ALGORITHM 67
A(20)./A(19) = [ 1.0000 1.0000 1.0000 -1.0000]
[ 0.4000 1.0000 0.4993 -1.0000]
[ 0.4000 0.4753 1.0000 -1.0000]
[ -0.2000 -0.5000 -0.5000 1.0000]

Notice that (4.2) does not state a rate for the element at position (3, 2).
These little numerical tests are intended to demonstrate that the convergence rates
given in (4.2) are in fact seen in a real run of the basic QR algorithm. The conclusions we
can draw are the following:

1. The convergence of the algorithm is slow. In fact it can be arbitrarily slow if eigen-
values are very close to each other.

2. The algorithm is expensive. Each iteration step requires the computation of the QR
factorization of a full n × n matrix, i.e., each single iteration step has a complexity
O(n3 ). Even if we assume that the number of steps is proportional to n we would
get an O(n4 ) complexity. The latter assumption is not even assured, see point 1 of
this discussion.

In the following we want to improve on both issues. First we want to find a matrix
structure that is preserved by the QR algorithm and that lowers the cost of a single
iteration step. Then, we want to improve on the convergence properties of the algorithm.

4.2 The Hessenberg QR algorithm

A matrix structure that is close to upper triangular form and that is preserved by the QR
algorithm is the Hessenberg form.

Definition 4.1 A matrix H is a Hessenberg matrix if its elements below the lower off-
diagonal are zero,
hij = 0, i > j + 1.

Theorem 4.2 The Hessenberg form is preserved by the QR algorithms.

Proof. We give a constructive proof, i.e., given a Hessenberg matrix H with QR factor-
ization H = QR, we show that H = RQ is again a Hessenberg matrix.
The Givens rotation or plane rotation G(i, j, ϑ) is defined by
 
1 ··· 0 ··· 0 ··· 0
 .. . . .. .. .. 
 . . . . . 
 
 0 ··· c ··· s ··· 0 
  ← i
 .. 
G(i, j, ϑ) :=  ... ..
.
..
. .
..
. 
 
(4.4)  0 · · · −s · · · c · · · 0  ← j
 
 .. .. .. . . .. 
 . . . . . 
0 ··· 0 ··· 0 ··· 1
↑ ↑
i j

where c = cos(ϑ) and s = sin(ϑ). Pre-multiplication by G(i, j, ϑ) amounts to a counter-

clockwise rotation by ϑ radians in the (i, j) coordinate plane. Clearly, a Givens rotation is
68 CHAPTER 4. THE QR ALGORITHM
an orthogonal matrix. For a unitary version see [4]. If x ∈ Rn and y = G(i, j, ϑ)∗ x, then

 cxi − sxj , k=i
yk = sxi + cxj , k=j

xk , k 6= i, j

We can force yj to be zero by setting

xi −xj
(4.5) c= p , s= p .
|xi + |xj |2
|2 |xi |2 + |xj |2

Thus, it is a simple matter to zero a single specific entry in a vector by using a Givens
rotation1 .
Now, let us look at a Hessenberg matrix H. We can show the principle procedure by
means of a 4 × 4 example.
   
× × × × × × × ×
× × × × G(1, 2, ϑ1 )∗ ·  0 × × ×
H=  
 0 × × × −−−−−−−−−→  0 × × ×


0 0 × × 0 0 × ×
   
× × × × × × × ×
G(2, 3, ϑ2 )∗ ·  0 × × × G(3, 4, ϑ3 )∗ ·  
 0 × × × = R
−−−−−−−−−→   0 0 × ×
− −− − −− −− −→  0 0 × ×
0 0 × × 0 0 0 ×

So, with Gk = G(k, k + 1, ϑk ), we get

G∗ G∗ G∗ H = R ⇐⇒ H = QR.
| 3 {z2 }1
Q∗

Multiplying Q and R in reversed order gives

H = RQ = RG1 G2 G3 ,

or, pictorially,
   
× × × × × × × ×
 0 × × × ·G(1, 2, ϑ1 ) × × × ×
R=  
 0 0 × × −−−−−−−−→  0 0 × ×


0 0 0 × 0 0 0 ×
   
× × × × × × × ×
·G(2, 3, ϑ2 )  × × × × ·G(3, 4, ϑ 1 ) 
× × × ×
−−−−−−−−→   0 × × × −−−−−−−−→  0 ×
=H
× ×
0 0 0 × 0 0 × ×

More generally, if H is n × n, n − 1 Givens rotations G1 , . . . , Gn−1 are needed to transform

H to upper triangular form. Applying the rotations from the right restores the Hessenberg
form.
Remark 4.1. The Hessenberg nonzero pattern isn’t the only pattern that is preserved by
the QR algoritm, see [2], however it is the most simple one.
1
For a stable way to compute Givens rotations see Algorithm 5.1.3 in [6].
4.2. THE HESSENBERG QR ALGORITHM 69
4.2.1 A numerical experiment
We repeat one of the previous two Matlab experiments

D = diag([4 3 2 1]);
rand(’seed’,0);
S=rand(4); S = (S - .5)*2;
A = S*D/S % A_0 = A = S*D*S^{-1}
H = hess(A); % built-in MATLAB function: generates
% unitarily similar Hessenberg matrix
for i=1:30,
[Q,R] = qr(H); H = R*Q
end
This yields the matrix sequence
H( 0) = [ -4.4529e-01 -1.8641e+00 -2.8109e+00 7.2941e+00]
[ 8.0124e+00 6.2898e+00 1.2058e+01 -1.6088e+01]
[ 0.0000e+00 4.0087e-01 1.1545e+00 -3.3722e-01]
[ 0.0000e+00 0.0000e+00 -1.5744e-01 3.0010e+00]

H( 5) = [ 4.0763e+00 -2.7930e+00 -7.1102e+00 2.1826e+01]

[ 5.6860e-02 2.4389e+00 -1.2553e+00 -3.5061e+00]
[ -2.0209e-01 2.5681e+00 -2.1805e+00]
[ 4.3525e-02 9.1667e-01]

H(10) = [ 4.0024e+00 -6.2734e-01 -7.0227e+00 -2.1916e+01]

[ 7.6515e-03 2.9123e+00 -9.9902e-01 3.3560e+00]
[ -8.0039e-02 2.0877e+00 3.3549e+00]
[ -7.1186e-04 9.9762e-01]

H(15) = [ 4.0001e+00 -1.0549e-01 -7.0411e+00 2.1902e+01]

[ 1.6833e-03 2.9889e+00 -8.9365e-01 -3.2181e+00]
[ -1.2248e-02 2.0111e+00 -3.6032e+00]
[ 2.0578e-05 9.9993e-01]

H(20) = [ 4.0000e+00 -3.1163e-02 -7.0425e+00 -2.1898e+01]

[ 3.9562e-04 2.9986e+00 -8.7411e-01 3.2072e+00]
[ -1.6441e-03 2.0014e+00 3.6377e+00]
[ -6.3689e-07 1.0000e-00]

H(25) = [ 4.0000e+00 -2.1399e-02 -7.0428e+00 2.1897e+01]

[ 9.3764e-05 2.9998e+00 -8.7056e-01 -3.2086e+00]
[ -2.1704e-04 2.0002e+00 -3.6423e+00]
[ 1.9878e-08 1.0000e-00]

H(30) = [ 4.0000e+00 -2.0143e-02 -7.0429e+00 -2.1897e+01]

[ 2.2247e-05 3.0000e+00 -8.6987e-01 3.2095e+00]
[ -2.8591e-05 2.0000e+00 3.6429e+00]
[ -6.2108e-10 1.0000e-00]

Finally we compute the element-wise quotients of the last two matrices.

H(30)./H(29) = [ 1.0000 0.9954 1.0000 -1.0000]
[ 0.7500 1.0000 0.9999 -1.0000]
[ 0.6667 1.0000 -1.0000]
[ -0.5000 1.0000]

Again the elements in the lower off-diagonal reflect nicely the convergence rates in (4.2).
70 CHAPTER 4. THE QR ALGORITHM
4.2.2 Complexity
We give the algorithm for a single Hessenberg-QR-step in a Matlab-like way, see Algo-
rithm 4.2. By
Hk:j,m:n ∈ C(j−k+1)×(n−m+1)

we denote the submatrix of H consisting of rows k through j and columns m through n.

Algorithm 4.2 A Hessenberg QR step

1: Let H ∈ Cn×n be an upper Hessenberg matrix. This algorithm overwrites H with
H = RQ where H = QR is a QR factorization of H.
2: for k = 1, 2, . . . , n − 1 do
3: /* Generate Gk and then apply it: H = G(k, k+1, ϑk )∗ H */
4: [ck , sk ] := givens(H
 k,k, Hk+1,k );
ck −sk
5: Hk:k+1,k:n = Hk:k+1,k:n;
sk ck
6: end for
7: for k = 1, 2, . . . , n − 1 do
8: /* Apply the rotations Gk from the right */
ck sk
9: H1:k+1,k:k+1 = H1:k+1,k:k+1 ;
−sk ck
10: end for

If we neglect the determination of the parameters ck and sk , see (4.5), then each of
the two loops requires
n−1
X n(n − 1)
6i = 6 ≈ 3n2 flops.
2
i=1

A flop is a floating point operation (+, −, ×, /). We do not distinguish between them,
although they may slightly differ in their execution time on a computer. Optionally, we
also have to execute the operation Uk := Uk−1 Qk of Algorithm 4.1. This is achieved by a
loop similar to the second loop in Algorithm 4.2. Since all the rows and columns of U are

1: for k=1,2,. . . ,n-1 do  

ck sk
2: U1:n,k:k+1 = U1:n,k:k+1 ;
−sk ck
3: end for

involved, executing the loop costs

n−1
X
6n ≈ 6n2 flops.
i=1

Altogether, a QR step with a Hessenberg matrix, including the update of the unitary
transformation matrix, requires 12n2 floating point operations. This has to be set in
relation to a QR step with a full matrix that costs 37 n3 . Consequently, we have gained a
factor of O(n) in terms of operations by moving from dense to Hessenberg form. However,
we may still have very slow convergence if one of the quotients |λk |/|λk+1 | is close to 1.
4.3. THE HOUSEHOLDER REDUCTION TO HESSENBERG FORM 71
4.3 The Householder reduction to Hessenberg form
In the previous section we discovered that it is a good idea to perform the QR algorithm
with Hessenberg matrices instead of full matrices. But we have not discussed how we
transform a full matrix (by means of similarity transformations) into Hessenberg form.
We catch up on this issue in this section.

4.3.1 Householder reflectors

Givens rotations are designed to zero a single element in a vector. Householder reflectors
are more efficient if a number of elements of a vector are to be zeroed at once. Here, we
follow the presentation given in [6].
Definition 4.3 A matrix of the form

P = I − 2uu∗ , kuk = 1,

is called a Householder reflector.

It is easy to verify that Householder reflectors are Hermitian and that P 2 = I. From this
we deduce that P is unitary. It is clear that we only have to store the Householder
vector u to be able to multiply a vector (or a matrix) with P ,

(4.6) P x = x − u(2u∗ x).

This multiplication only costs 4n flops where n is the length of the vectors.
A task that we repeatedly want to carry out with Householder reflectors is to transform
a vector x on a multiple of e1 ,

P x = x − u(2u∗ x) = αe1 .

Since P is unitary, we must have α = ρkxk, where ρ ∈ C has absolute value one. Therefore,
 
x1 − ρkxk
x − ρkxke1 1  x2 
 
u= =  .. 
kx − ρkxke1 k kx − ρkxke1 k  . 
xn

We can freely choose ρ provided that |ρ| = 1. Let x1 = |x1 |eiφ . To avoid numerical
cancellation we set ρ = −eiφ .
In the real case, one commonly sets ρ = −sign(x1 ). If x1 = 0 we can set ρ in any way.

4.3.2 Reduction to Hessenberg form

Now we show how to use Householder reflectors to reduce an arbitrary square matrix to
Hessenberg form. We show the idea by means of a 5 × 5 example. In the first step of the
reduction we introduce zeros in the first column below the second element,
     
× × × × × × × × × × × × × × ×
× × × × ×    
  P1 ∗ × × × × × ∗P1 × × × × ×
A=    
× × × × × −−−→  0 × × × × −−−→  0 × × × × = P1 AP1 .
 ∗
× × × × ×  0 × × × ×  0 × × × ×
× × × × × 0 × × × × 0 × × × ×
72 CHAPTER 4. THE QR ALGORITHM
Notice that P1 = P1∗ since it is a Householder reflector! It has the structure
 
1 0 0 0 0
0 × × × ×  
  1 0T
P1 = 
0 × × × 
× = .
0 0 I4 − 2u1 u∗1
× × × ×
0 × × × ×

The Householder vector u1 is determined such that

     
a21 α u1
 a    u2 
(I − 2u1 u∗1 )  31  = 0 with u1 =  
a41   0  u3  .
a51 0 u4

The multiplication of P1 from the left inserts the desired zeros in column 1 of A. The
multiplication from the right is necessary in order to have similarity. Because of the
nonzero structure of P1 the first column of P1 A is not affected. Hence, the zeros stay
there.
The reduction continues in a similar way:
   
× × × × × × × × × ×
× × × × ×  
  P2 ∗ / ∗ P2 × × × × ×
 
P1 AP1 =  0 × × × × −−−−−−−−→  0 × × × ×

 0 × × × ×  0 0 × × ×
0 × × × × 0 0 × × ×
 
× × × × ×
× × × × ×
P3 ∗ / ∗ P3  
−−−−−−−−→   0 × × × ×  = P3 P2 P1 A P1 P2 P3 .

 0 0 × × × | {z }
U
0 0 0 × ×

Algorithm 4.3 gives the details for the general n × n case. In step 4 of this algorithm,
the Householder reflector is generated such that
     
ak+1,k α u1
ak+2,k     u2 
  0  
(I − 2uk u∗k )  .  =    with u k =  ..  and |α| = kxk
.
 .  0  . 
an,k 0 un−k

according to the considerations of the previous subsection. The Householder vectors are
stored at the locations of the zeros. Therefore the matrix U = P1 · · · Pn−2 that effects the
similarity transformation from the full A to the Hessenberg H is computed after all House-
holder vectors have been generated, thus saving (2/3)n3 flops. The overall complexity of
the reduction is
P
n−2
• Application of Pk from the left: 4(n − k − 1)(n − k) ≈ 34 n3
k=1

n−2
P
• Application of Pk from the right: 4(n)(n − k) ≈ 2n3
k=1
4.4. IMPROVING THE CONVERGENCE OF THE QR ALGORITHM 73
Algorithm 4.3 Reduction to Hessenberg form
1: This algorithm reduces a matrix A ∈ Cn×n to Hessenberg form H by a sequence of
Householder reflections. H overwrites A.
2: for k = 1 to n−2 do
3: Generate the Householder reflector Pk ;
4: /* Apply Pk = Ik ⊕ (In−k − 2uk uk ∗ ) from the left to A */
5: Ak+1:n,k:n := Ak+1:n,k:n − 2uk (uk ∗ Ak+1:n,k:n);
6: /* Apply Pk from the right, A := APk */
7: A1:n,k+1:n := A1:n,k+1:n − 2(A1:n,k+1:n uk )uk ∗ ;
8: end for
9: if eigenvectors are desired form U = P1 · · · Pn−2 then
10: U := In ;
11: for k = n−2 downto 1 do
12: /* Update U := Pk U */
13: Uk+1:n,k+1:n := Uk+1:n,k+1:n − 2uk (uk ∗ Uk+1:n,k+1:n);
14: end for
15: end if

P
n−2
• Form U = P1 · · · Pn−2 : 4(n − k)(n − k) ≈ 43 n3
k=1

Thus, the reduction to Hessenberg form costs 10 3

3 n flops without forming the transforma-
14 3
tion matrix and 3 n including forming this matrix.

4.4 Improving the convergence of the QR algorithm

We have seen how the QR algorithm for computing the Schur form of a matrix A can be
executed more economically if the matrix A is first transformed to Hessenberg form. Now
we want to show how the convergence of the Hessenberg QR algorithm can be improved
dramatically by introducing (spectral) shifts into the algorithm.

Lemma 4.4 Let H be an irreducible Hessenberg matrix, i.e., hi+1,i 6= 0 for all i =
1, . . . , n − 1. Let H = QR be the QR factorization of H. Then for the diagonal elements
of R we have
|rkk | > 0, for all k < n.
Thus, if H is singular then rnn = 0.

Proof. Let us look at the k-th step of the Hessenberg QR factorization. For illustration,
let us consider the case k = 3 in a 5 × 5 example, where the matrix has the structure
 
+ + + + +
 0 + + + +
 
 0 0 + + + .
 
 0 0 × × ×
0 0 0 × ×
The plus-signs indicate elements that have been modified. In step 3, the (nonzero) element
h43 will be zeroed by a Givens rotation G(3, 4, ϕ) that is determined such that
    
cos(ϕ) − sin(ϕ) h̃kk r
= kk .
sin(ϕ) cos(ϕ) hk+1,k 0
74 CHAPTER 4. THE QR ALGORITHM
Because the Givens rotation preserves vector lengths, we have

|rkk |2 = |h̃kk |2 + |hk+1,k |2 ≥ |hk+1,k |2 > 0,

which confirms the claim.

We apply this Lemma to motivate a further strategy to speed up the convergence of
the QR algorithm.
Let λ be an eigenvalue of the irreducible Hessenberg matrix H. Let us check what
happens it we perform

1: H − λI = QR /* QR factorization */
2: H = RQ + λI

First we notice that H ∼ H. In fact,

H = Q∗ (H − λI)Q + λI = Q∗ HQ.

Second, by Lemma 4.4 we have

" #
H − λI = QR, with R= .
0

Thus, " #
RQ =
00

and " # " #

H 1 h1
H = RQ + λI = = .
0λ 0T λ
So, if we apply a QR step with a perfect shift to a Hessenberg matrix, the eigenvalue
drops out. We then could deflate, i.e., proceed the algorithm with the smaller matrix
H 1.
Remark 4.2. We could prove the existence of the Schur decomposition in the following
way. (1) transform the arbitrary matrix to Hessenberg form. (2) Do the perfect shift
Hessenberg QR with the eigenvalues which we known to exist one after the other.

4.4.1 A numerical example

We use a matrix of a previous Matlab experiments to show that perfect shifts actually
work.

D = diag([4 3 2 1]); rand(’seed’,0);

S=rand(4); S = (S - .5)*2;
A = S*D/S;
format short e
H = hess(A)
[Q,R] = qr(H - 2*eye(4))
H1 = R*Q + 2*eye(4)
format long
lam = eig(H1(1:3,1:3))
4.4. IMPROVING THE CONVERGENCE OF THE QR ALGORITHM 75
Matlab produces the output
H = [ -4.4529e-01 -1.8641e+00 -2.8109e+00 7.2941e+00]
[ 8.0124e+00 6.2898e+00 1.2058e+01 -1.6088e+01]
[ 4.0087e-01 1.1545e+00 -3.3722e-01]
[ -1.5744e-01 3.0010e+00]

Q = [ -2.9190e-01 -7.6322e-01 -4.2726e-01 -3.8697e-01]

[ 9.5645e-01 -2.3292e-01 -1.3039e-01 -1.1810e-01]
[ 6.0270e-01 -5.9144e-01 -5.3568e-01]
[ -6.7130e-01 7.4119e-01]

R = [ 8.3772e+00 4.6471e+00 1.2353e+01 -1.7517e+01]

[ 6.6513e-01 -1.1728e+00 -2.0228e+00]
[ 2.3453e-01 -1.4912e+00]
[ -2.4425e-14]

H1 = [ 3.9994e+00 -3.0986e-02 2.6788e-01 -2.3391e+01]

[ 6.3616e-01 1.1382e+00 1.9648e+00 -9.4962e-01]
[ 1.4135e-01 2.8623e+00 -1.2309e+00]
[ 1.6396e-14 2.0000e+00]

lam = [9.99999999999993e-01 4.00000000000003e+00 3.00000000000000e+00]

4.4.2 QR algorithm with shifts

This considerations indicate that it may be good to introduce shifts into the QR algorithm.
However, we cannot choose perfect shifts because we do not know the eigenvalues of the
matrix! We therefore need heuristics how to estimate eigenvalues. One such heuristic is
the Rayleigh quotient shift: Set the shift σk in the k-th step of the QR algorithm equal
to the last diagonal element:

(4.7) σk := h(k−1)
n,n = e∗n H (k−1) en .

Algorithm 4.4 The Hessenberg QR algorithm with Rayleigh quotient shift

1: Let H0 = H ∈ Cn×n be an upper Hessenberg matrix. This algorithm computes its
Schur normal form H = U T U ∗ .
2: k := 0;
3: for m=n,n-1,. . . ,2 do
4: repeat
5: k := k + 1;
(k−1)
6: σk := hm,m ;
7: Hk−1 − σk I =: Qk Rk ;
8: Hk := Rk Qk + σk I;
9: Uk := Uk−1 Qk ;
(k)
10: until |hm,m−1 | is sufficiently small
11: end for
12: T := Hk ;

Algorithm 4.4 implements this heuristic. Notice that the shift changes in each iteration
step! Notice also that deflation is incorporated in Algorithm 4.4. As soon as the last lower
off-diagonal element is sufficiently small, it is declared zero, and the algorithm proceeds
with a smaller matrix. In Algorithm 4.4 the ‘active portion’ of the matrix is m × m.
76 CHAPTER 4. THE QR ALGORITHM
Lemma 4.4 guarantees that a zero is produced at position (n, n − 1) in the Hessenberg
matrix H if the shift equals an eigenvalue of H. What happens, if hn,n is a good approx-
imation to an eigenvalue of H? Let us assume that we have an irreducible Hessenberg
matrix  
× × × × ×
× × × × × 
 
0 × × × × ,
 
0 0 × × × 
0 0 0 ε hn,n
where ε is a small quantity. If we perform a shifted Hessenberg QR step, we first have to
factor H − hn,n I, QR = H − hn,n I. After n − 2 steps of this factorization the R-factor is
almost upper triangular,  
+ + + + +
 0 + + + +
 
 0 0 + + + .
 
0 0 0 α β
0 0 0 ε 0
From (4.5) we see that the last Givens rotation has the nontrivial elements
α −ε
cn−1 = p , sn−1 = p .
|α|2 + |ε|2 |α|2 + |ε|2

Applying the Givens rotations from the right one sees that the last lower off-diagonal
element of H = RQ + hn,n I becomes

ε2 β
(4.8) h̄n,n−1 = .
α2 + ε2
So, we have quadratic convergence unless α is also tiny.
A second even more often used shift strategy is the Wilkinson shift:
" (k−1) (k−1)
#
hn−1,n−1 hn−1,n
(4.9) σk := eigenvalue of (k−1) (k−1)
that is closer to h(k−1)
n,n .
hn,n−1 hn,n

4.4.3 A numerical example

We give an example for the Hessenberg QR algorithm with shift, but without deflation.
The Matlab code

D = diag([4 3 2 1]);
rand(’seed’,0);
S=rand(4); S = (S - .5)*2;
A = S*D/S;
H = hess(A)

for i=1:8,
[Q,R] = qr(H-H(4,4)*eye(4)); H = R*Q+H(4,4)*eye(4);
end
produces the output
4.5. THE DOUBLE SHIFT QR ALGORITHM 77
H( 0) = [ -4.4529e-01 -1.8641e+00 -2.8109e+00 7.2941e+00]
[ 8.0124e+00 6.2898e+00 1.2058e+01 -1.6088e+01]
[ 0.0000e+00 4.0087e-01 1.1545e+00 -3.3722e-01]
[ 0.0000e+00 0.0000e+00 -1.5744e-01 3.0010e+00]

H( 1) = [ 3.0067e+00 1.6742e+00 -2.3047e+01 -4.0863e+00]

[ 5.2870e-01 8.5146e-01 1.1660e+00 -1.5609e+00]
[ -1.7450e-01 3.1421e+00 -1.1140e-01]
[ -1.0210e-03 2.9998e+00]

H( 2) = [ 8.8060e-01 -4.6537e-01 9.1630e-01 1.6146e+00]

[ -1.7108e+00 5.3186e+00 2.2839e+01 -4.0224e+00]
[ -2.2542e-01 8.0079e-01 5.2445e-01]
[ -1.1213e-07 3.0000e+00]

H( 3) = [ 1.5679e+00 9.3774e-01 1.5246e+01 1.2703e+00]

[ 1.3244e+00 2.7783e+00 1.7408e+01 4.1764e+00]
[ 3.7230e-02 2.6538e+00 -7.8404e-02]
[ 8.1284e-15 3.0000e+00]

H( 4) = [ 9.9829e-01 -7.5537e-01 -5.6915e-01 1.9031e+00]

[ -3.2279e-01 5.1518e+00 2.2936e+01 -3.9104e+00]
[ -1.6890e-01 8.4993e-01 3.8582e-01]
[ -5.4805e-30 3.0000e+00]

H( 5) = [ 9.3410e-01 -3.0684e-01 3.0751e+00 -1.2563e+00]

[ 3.5835e-01 3.5029e+00 2.2934e+01 4.1807e+00]
[ 3.2881e-02 2.5630e+00 -7.2332e-02]
[ 1.1313e-59 3.0000e+00]

H( 6) = [ 1.0005e+00 -8.0472e-01 -8.3235e-01 1.9523e+00]

[ -7.5927e-02 5.1407e+00 2.2930e+01 -3.8885e+00]
[ -1.5891e-01 8.5880e-01 3.6112e-01]
[ -1.0026e-119 3.0000e+00]

H( 7) = [ 9.7303e-01 -6.4754e-01 -8.9829e-03 -1.8034e+00]

[ 8.2551e-02 3.4852e+00 2.3138e+01 3.9755e+00]
[ 3.3559e-02 2.5418e+00 -7.0915e-02]
[ 3.3770e-239 3.0000e+00]

H( 8) = [ 1.0002e+00 -8.1614e-01 -8.9331e-01 1.9636e+00]

[ -1.8704e-02 5.1390e+00 2.2928e+01 -3.8833e+00]
[ -1.5660e-01 8.6086e-01 3.5539e-01]
[ 0 3.0000e+00]
The numerical example shows that the shifted Hessenberg QR algorithm can work very
nicely. In this example the (4,3) element is about 10−30 after 3 steps. (We could stop
there.) The example also nicely shows a quadratic convergence rate.

4.5 The double shift QR algorithm

The shifted Hessenberg QR algorithm does not always work so nicely as in the previous
example. If α in (4.8) is O(ε) then hn,n−1 can be large. (A small α indicates a near
singular H1:n−1,1:n−1.)
Another problem occurs if real Hessenberg matrices have complex eigenvalues. We
know that for reasonable convergence rates the shifts must be complex. If an eigenvalue
λ has been found we can execute a single perfect shift with λ̄. It is (for rounding errors)
unprobable however that we will get back to a real matrix.
78 CHAPTER 4. THE QR ALGORITHM
Since the eigenvalues come in complex conjugate pairs it is natural to search for a pair
of eigenvalues right-away. This is done by collapsing two shifted QR steps in one double
step with the two shifts being complex conjugates of each other.
Let σ1 and σ2 be two eigenvalues of the real matrix (cf. Wilkinson shift (4.9))
" (k−1) (k−1)
#
hn−1,n−1 hn−1,n
G= (k−1) (k−1)
∈ R2×2 .
hn,n−1 hn,n

If σ1 ∈ C \ R then σ2 = σ̄1 . Let us perform two QR steps using σ1 and σ2 as shifts. Setting
k = 1 for convenience we get

H 0 − σ 1 I = Q 1 R1 ,
H1 = R1 Q1 + σ1 I,
(4.10)
H 1 − σ 2 I = Q 2 R2 ,
H2 = R2 Q2 + σ2 I.

From the second and third equation in (4.10) we obtain

R1 Q1 + (σ1 − σ2 )I = Q2 R2 .

Multiplying this equation with Q1 from the left and with R1 from the right we get

Q1 R1 Q1 R1 + (σ1 − σ2 )Q1 R1 = Q1 R1 (Q1 R1 + (σ1 − σ2 )I)

= (H0 − σ1 I)(H0 − σ2 I) = Q1 Q2 R2 R1 .

Because σ2 = σ̄1 we have

M := (H0 − σ1 I)(H0 − σ̄1 I) = H02 − 2Re(σ)H0 + |σ|2 I = Q1 Q2 R2 R1 .

Therefore, (Q1 Q2 )(R2 R1 ) is the QR factorization of a real matrix. We can choose (scale)
Q1 and Q2 such that Z := Q1 Q2 is real orthogonal. (Then also R2 R1 is real.) By
consequence,
H2 = (Q1 Q2 )∗ H0 (Q1 Q2 ) = Z T H0 Z
is real.
A procedure to compute H2 by avoiding complex arithmetic could consist of three
steps:
(k−1)
1. Form the real matrix M = H02 − sH0 + tI with s = 2Re(σ) = trace(G) = hn−1,n−1 +
(k−1) (k−1) (k−1) (k−1) (k−1)
hn,n and t = |σ|2 = det(G) = hn−1,n−1 hn,n − hn−1,n hn,n−1 . Notice that M has
two lower off-diagonals, " #
M= .

2. Compute the QR factorization M = ZR,

3. Set H2 = Z T H0 Z.

This procedure is however too expensive since item 1, i.e., forming H 2 requires O(n3 )
flops.
A remedy for the situation is provided by the Implicit Q Theorem.
4.5. THE DOUBLE SHIFT QR ALGORITHM 79
Theorem 4.5 (The implicit Q theorem) Let A ∈ Rn×n . Let Q = [q1 , . . . , qn ] and
V = [v1 , . . . , vn ] be orthogonal matrices that both similarly transform A to Hessenberg
form, H = QT AQ and G = V T AV . Let k denote the smallest positive integer for which
hk+1,k = 0, with k = n if H is irreducible.
If q1 = v1 then qi = ±vi and |hi,i−1 | = |gi,i−1 | for i = 2, . . . , k. If k < n, then
gk+1,k = 0.

Proof. [6] Let W = V T Q. Clearly, W is orthogonal, and GW = W H.

We first show that the first k columns of W form an upper triangular matrix, i.e.,

(4.11) wi = W ei ∈ span{e1 , . . . , ei }, i ≤ k.

(Notice that orthogonal upper triangular matrices are diagonal with diagonal entries ±1.)
This is proced inductively. For i = 1 we have w1 = e1 by the assumption that q1 = v1 .
For 1 < i ≤ k we assume that (4.11) is true for wi and use the equality GW = W H. The
(i−1)-th column of this equation reads

i
X
Gwi−1 = GW ei−1 = W Hei−1 = wj hj,i−1 .
j=1

Since hi,i−1 6= 0 we have

i−1
X
wi hi,i−1 = Gwi−1 − wj hj,i−1 ∈ span{e1 , . . . ei },
j=1

as G is a Hessenberg matrix. So, the upper-left k × k block of W is upper triangular.

Since the columns of W are orthogonal we conclude that wi = ±ei , i ≤ k.
Since wi = ±V T Qei = V T qi = ±ei we see that qi is orthogonal to all columns of V
except the i-th. Therefore, we must have qi = ±vi . Further,

hi,i−1 = eTi Hei−1 = eTi QT AQei−1 = eTi QT V GV T Qei−1 = wiT Gwi−1 = ±gi,i−1 ,

thus, |hi,i−1 | = |gi,i−1 |. If hk+1,k = 0 then

k
X
gk+1,k = eTk+1 Gek = ± eTk+1 GW ek = ± eTk+1 W Hek = ± eTk+1 wj hj,k = 0.
j=1

since eTk+1 wj = ±eTk+1 ej = 0 for j ≤ k.

Golub and van Loan [6, p.347] write that “The gist of the implicit Q theorem is that if
QT AQ = H and Z T AZ = G are both unreduced Hessenberg matrices and Q and Z have
the same first column, then G and H are “essentially equal” in the sense that G = DHD
with D = diag(±1, . . . , ±1).”

We apply the Implicit Q Theorem in the following way: We want to compute the
Hessenberg matrix Hk+1 = Z T Hk−1 Z where ZR is the QR factorization of M = Hk−1 2 −
sHk−1 + tI. The Implicit Q Theorem now tells us that we essentially get Hk+1 by any
orthogonal similarity transformation Hk−1 → Z1∗ Hk−1 Z1 provided that Z1∗ HZ1 is Hessen-
berg and Z1 e1 = Ze1 .
80 CHAPTER 4. THE QR ALGORITHM
Let P0 be the Householder reflector with

P0T M e1 = P0T (Hk−1

2
− 2Re(σ)Hk−1 + |σ|2 I) e1 = α e1 .

Since only the first three elements of the first column M e1 of M are nonzero, P0 has the
structure  
× × ×
× × × 
 
× × × 
 
P0 =  1 .
 
 . 
 . . 
1
So,  
× × × × × × ×
 × × × × × × × 
 
 + × × × × × × 
 
′
Hk−1 := P0 Hk−1 P0 = 
T
 + + × × × × × .

 × × × × 
 
 × × × 
× ×
We now reduce P0T Hk−1 P0 similarly to Hessenberg form the same way as we did earlier, by
a sequence of Householder reflectors P1 , . . . , Pn−2 . However, P0T Hk−1 P0 is a Hessenberg
matrix up to the bulge at the top left. We take into account this structure when forming
the Pi = I − 2pi pTi . So, the structures of P1 and of P1T P0T Hk−1 P0 P1 are
   
1 × × × × × × ×
 × × ×   × × × × × × × 
   
 × × ×   0 × × × × × × 
   
P1 = 
 × × × , H ′′ = P T H ′ P1 =  0 + × × × × × .
 k−1 1 k−1  
 1   + + × × × × 
   
 1   × × × 
1 × ×

The transformation with P1 has chased the bulge one position down the diagonal. The
consecutive reflectors push it further by one position each until it falls out of the matrix
at the end of the diagonal. Pictorially, we have
 
× × × × × × ×
 × × × × × × × 
 
 × × × × × × 
 
′′′
Hk−1 = P2T Hk−1
′′
P2 = 
 0 × × × × ×  
 0 + × × × × 
 
 + + × × × 
 
× × × × × × × × ×
 × × × × × × × 
 
 × × × × × × 
 
Hk−1 = P3 Hk−1 P3 = 
′′′′ T ′′′
 × × × × × 

 0 × × × × 
 
 0 + × × × 
+ + × ×
4.5. THE DOUBLE SHIFT QR ALGORITHM 81

 
× × × × × × ×
 × × × × × × × 
 
 × × × × × × 
 
′′′′′
Hk−1 = P4T Hk−1
′′′′
P4 = 
 × × × × × 

 × × × × 
 
 0 × × ×   
× × × × × × ×
0 + × ×  × × × 
 × × × × 
 × × × × × × 
 
′′′′′′
Hk−1 T ′′′′′ 
= P5 Hk−1 P5 =  × × × × × 

 × × × × 
 
 × × × 
0 × ×

It is easy to see that the Householder vector pi , i < n − 2, has only three nonzero elements
at position i + 1, i + 2, i + 3. Of pn−2 only the last two elements are nonzero. Clearly,
P0 P1 · · · Pn−2 e1 = P0 e1 = M e1 /α.
Remark 4.3. Notice that in Algorithm 4.5 a double step is taken also if the eigenvalues of
 
hqq hqp
G=
hpq hpp

are real. As in the complex case we set s = trace(G) and t = det(G).

4.5.1 A numerical example

We consider a simple Matlab implementation of the Algorithm 4.5 to compute the eigen-
values of the real matrix
 
7 3 4 −11 −9 −2
 −6 4 −5 7 1 12 
 
 −1 −9 2 2 9 1 
A=
 −8

 0 −1 5 0 8 

 −4 3 −5 7 2 10 
6 1 4 −11 −7 −1

that has the spectrum

σ(A) = {1 ± 2i, 3, 4, 5 ± 6i}.
The intermediate output of the code was (after some editing) the following:
>> H=hess(A)

H(0) =

7.0000 7.2761 5.8120 -0.1397 9.0152 7.9363

12.3693 4.1307 18.9685 -1.2071 10.6833 2.4160
0 -7.1603 2.4478 -0.5656 -4.1814 -3.2510
0 0 -8.5988 2.9151 -3.4169 5.7230
0 0 0 1.0464 -2.8351 -10.9792
0 0 0 0 1.4143 5.3415
82 CHAPTER 4. THE QR ALGORITHM
Algorithm 4.5 The Francis double step QR algorithm
1: Let H0 = H ∈ Rn×n be an upper Hessenberg matrix. This algorithm computes its
real Schur form H = U T U T using the Francis double step QR algorithm. T is a quasi
upper triangular matrix.
2: p := n; /* p indicates the ‘active’ matrix size. */
3: while p > 2 do
4: q := p − 1;
5: s := Hq,q + Hp,p; t := Hq,q Hp,p − Hq,p Hp,q ;
6: /* compute first 3 elements of first column of M */
2 +H H
7: x := H1,1 1,2 2,1 − sH1,1 + t;
8: y := H2,1 (H1,1 + H2,2 − s);
9: z := H2,1 H3,2 ;
10: for k = 0 to p − 3 do
11: Determine the Householder reflector P with P T [x; y; z]T = αe1 ;
12: r := max{1, k};
13: Hk+1:k+3,r:n := P T Hk+1:k+3,r:n;
14: r := min{k + 4, p};
15: H1:r,k+1:k+3 := H1:r,k+1:k+3P ;
16: x := Hk+2,k+1 ; y := Hk+3,k+1 ;
17: if k < p − 3 then
18: z := Hk+4,k+1 ;
19: end if
20: end for
21: Determine the Givens rotation P with P T [x; y]T = αe1 ;
22: Hq:p,p−2:n := P T Hq:p,p−2:n;
23: H1:p,p−1:p := H1:p,p−1:pP ;
24: /* check for convergence */
25: if |Hp,q | < ε (|Hq,q | + |Hp,p |) then
26: Hp,q := 0; p := p − 1; q := p − 1;
27: else if |Hp−1,q−1 | < ε (|Hq−1,q−1 | + |Hq,q |) then
28: Hp−1,q−1 := 0; p := p − 2; q := p − 1;
29: end if
30: end while

>> PR=qr2st(H)

[it_step, p = n_true, H(p,p-1), H(p-1,p-2)]

1 6 -1.7735e-01 -1.2807e+00
2 6 -5.9078e-02 -1.7881e+00
3 6 -1.6115e-04 -5.2705e+00
4 6 -1.1358e-07 -2.5814e+00
5 6 1.8696e-14 1.0336e+01
6 6 -7.1182e-23 -1.6322e-01

H(6) =

5.0000 6.0000 2.3618 5.1837 -13.4434 -2.1391

-6.0000 5.0000 2.9918 10.0456 -8.7743 -21.0094
0.0000 -0.0001 -0.9393 3.6939 11.7357 3.8970
4.5. THE DOUBLE SHIFT QR ALGORITHM 83
0.0000 -0.0000 -1.9412 3.0516 2.9596 -10.2714
0 0.0000 0.0000 -0.1632 3.8876 4.1329
0 0 0 0.0000 -0.0000 3.0000

7 5 1.7264e-02 -7.5016e-01
8 5 2.9578e-05 -8.0144e-01
9 5 5.0602e-11 -4.6559e+00
10 5 -1.3924e-20 -3.1230e+00

H(10) =

5.0000 6.0000 -2.7603 1.3247 11.5569 -2.0920

-6.0000 5.0000 -10.7194 0.8314 11.8952 21.0142
-0.0000 -0.0000 3.5582 3.3765 5.9254 -8.5636
-0.0000 -0.0000 -3.1230 -1.5582 -10.0935 -6.3406
0 0 0 0.0000 4.0000 4.9224
0 0 0 0.0000 0 3.0000

11 4 1.0188e+00 -9.1705e-16

H(11) =

5.0000 6.0000 -10.2530 4.2738 -14.9394 -19.2742

-6.0000 5.0000 -0.1954 1.2426 7.2023 -8.6299
-0.0000 -0.0000 2.2584 -5.4807 -10.0623 4.4380
0.0000 -0.0000 1.0188 -0.2584 -5.9782 -9.6872
0 0 0 0 4.0000 4.9224
0 0 0 0.0000 0 3.0000

4.5.2 The complexity

We first estimate the complexity of a single step of the double step Hessenberg QR al-
gorithm. The most expensive operations are the applications of the 3 × 3 Householder
reflectors in steps 13 and 15 of Algorithm 4.5. Let us first count the flops for applying the
Householder reflector to a 3-vector,

x := (I − 2uuT )x = x − u(2uT x).

The inner product uT x costs 5 flops, multiplying with 2 another one. The operation
x := x − uγ, γ = 2uT x, cost 6 flops, altogether 12 flops.
In the k-th step of the loop there are n − k of these application from the left in step 13
and k + 4 from the right in step 15. In this step there are thus about 12n + O(1) flops to be
executed. As k is running from 1 to p − 3. We have about 12pn flops for this step. Since p
runs from n down to about 2 we have 6n3 flops. If we assume that two steps are required per
eigenvalue the flop count for Francis’ double step QR algorithm to compute all eigenvalues
of a real Hessenberg matrix is 12n3 . If also the eigenvector matrix is accumulated the two
additional statements have to be inserted into Algorithm 4.5. After step 15 we have

1: Q1:n,k+1:k+3 := Q1:n,k+1:k+3 P ;

and after step 23 we introduce

84 CHAPTER 4. THE QR ALGORITHM

1: Q1:n,p−1:p := Q1:n,p−1:pP ;
which costs another 12n3 flops.
We earlier gave the estimate of 6n3 flops for a Hessenberg QR step, see Algorithm 4.2.
If the latter has to be spent in complex arithmetic then the single shift Hessenberg QR al-
gorithm is more expensive than the double shift Hessenberg QR algorithm that is executed
in real arithmetic.
Remember that the reduction to Hessenberg form costs 10 3
3 n flops without forming the
14 3
transformation matrix and 3 n if this matrix is formed.

4.6 The symmetric tridiagonal QR algorithm

The QR algorithm can be applied straight to Hermitian or symmetric matrices. By (4.1) we
see that the QR algorithm generates a sequence {Ak } of symmetric matrices. Taking into
account the symmetry, the performance of the algorithm can be improved considerably.
Furthermore, from Theorem 2.14 we know that Hermitian matrices have a real spectrum.
Therefore, we can restrict ourselves to single shifts.

4.6.1 Reduction to tridiagonal form

The reduction of a full Hermitian matrix to Hessenberg form produces a Hermitian Hes-
senberg matrix, which (up to rounding errors) is a real symmetric tridiagonal matrix. Let
us consider how to take into account symmetry. To that end let us consider the first
reduction step that introduces n − 2 zeros into the first column (and the first row) of
A = A∗ ∈ Cn×n . Let
 
1 0T
P1 = , u1 ∈ Cn , ku1 k = 1.
0 In−1 − 2u1 u∗1
Then,
A1 := P1∗ AP1 = (I − 2u1 u∗1 )A(I − 2u1 u∗1 )
= A − u1 (2u∗1 A − 2(u∗1 Au1 )u∗1 ) − (2Au1 − 2u1 (u∗1 Au1 )) u∗1
| {z } | {z }
v1∗ v1
= A − u1 v1∗ − v1 u∗1 .
In the k-th step of the reduction we similarly have
Ak = Pk∗ Ak−1 Pk = Ak−1 − uk−1 vk−1
∗
− vk−1 u∗k−1 ,
where the last n − k elements of uk−1 and vk−1 are nonzero. Forming
vk−1 = 2Ak−1 uk−1 − 2uk−1 (u∗k−1 Ak−1 uk−1 )
costs 2(n − k)2 + O(n − k) flops. This complexity results from Ak−1 uk−1 . The rank-2
update of Ak−1 ,
∗
Ak = Ak−1 − uk−1 vk−1 − vk−1 u∗k−1 ,
requires another 2(n−k)2 +O(n−k) flops, taking into account symmetry. By consequence,
the transformation to tridiagonal form can be accomplished in
n−1
X
4
4(n − k)2 + O(n − k) = n3 + O(n2 )
3
k=1
4.6. THE SYMMETRIC TRIDIAGONAL QR ALGORITHM 85
floating point operations.

4.6.2 The tridiagonal QR algorithm

In the symmetric case the Hessenberg QR algorithm becomes a tridiagonal QR algorithm.
This can be executed in an explicit or an implicit way. In the explicit form, a QR step
is essentially

1: Choose a shift µ
2: Compute the QR factorization A − µI = QR
3: Update A by A = RQ + µI.

Of course, this is done by means of plane rotations and by respecting the symmetric
tridiagonal structure of A.
In the more elegant implicit form of the algorithm we first compute the first Givens
rotation G0 = G(1, 2, ϑ) of the QR factorization that zeros the (2, 1) element of A − µI,
    
c s a11 − µ ∗
(4.12) = , c = cos(ϑ0 ), s = sin(ϑ0 ).
−s c a21 0

Performing a similary transformation with G0 we have (n = 5)

 
× × +
× × × 
 
G∗0 AG0 = A′ = 
+ × × × 

 × × ×
× ×
Similar as with the double step Hessenberg QR algorithm we chase the bulge down the
diagonal. In the 5 × 5 example this becomes
   
× × + × × 0
× × ×  × × × + 
G0   G1  

A −−−−−−−−−−→ + × × ×   
 −−−−−−−−−−→  0 × × × 
= G(1, 2, ϑ0 )  × × × = G(2, 3, ϑ 1 )  + × × ×
× × × ×
   
× × 0 × ×
× × ×  × × × 
G2   G 3
 
−−−−−−−−−−→   × × × +  −−−−−−−−−−→ 
 × × × 0  = A.
= G(3, 4, ϑ2 )  0 × × ×  = G(4, 5, ϑ 3 )  × × ×
+ × × 0 × ×
The full step is given by

A = Q∗ AQ, Q = G0 G1 · · · Gn−2 .

Because Gk e1 = e1 for k > 0 we have

Q e1 = G0 G1 · · · Gn−2 e1 = G0 e1 .

Both explicit and implicit QR step form the same first plane rotation G0 . By referring to
the Implicit Q Theorem 4.5 we see that explicit and implicit QR step compute essentially
the same A.
86 CHAPTER 4. THE QR ALGORITHM

Algorithm 4.6 Symmetric tridiagonal QR algorithm with implicit Wilkinson

shift
1: Let T ∈ Rn×n be a symmetric tridiagonal matrix with diagonal entries a1 , . . . , an and
off-diagonal entries b2 , . . . , bn .
This algorithm computes the eigenvalues λ1 , . . . , λn of T and corresponding eigenvec-
tors q1 , . . . , qn . The eigenvalues are stored in a1 , . . . , an . The eigenvectors are stored
in the matrix Q, such that T Q = Q diag(a1 , . . . , an ).

2: m = n /* Actual problem dimension. m is reduced in the convergence check. */

3: Q = In ;
4: while m > 1 do
5: d := (am−1 − am )/2; /* Compute Wilkinson’s shift */
6: if d = 0 then
7: s := am − |bm |;
8: else p
9: s := am − b2m /(d + sign(d) d2 + b2m );
10: end if
11: x := a(1) − s; /* Implicit QR step begins here */
12: y := b(2);
13: for k = 1 to m − 1 do
14: if m > 2 then
15: [c, s] := givens(x, y);
16: else    
c −s a1 b2 c s
17: Determine [c, s] such that is diagonal
s c b2 a2 −s c
18: end if
19: w := cx − sy;
20: d := ak − ak+1 ; z := (2cbk+1 + ds)s;
21: ak := ak − z; ak+1 := ak+1 + z;
22: bk+1 := dcs + (c2 − s2 )bk+1 ;
23: x := bk+1 ;
24: if k > 1 then
25: bk := w;
26: end if
27: if k < m − 1 then
28: y := −sbk+2 ; bk+2 := cbk+2 ;
29: end if  
c s
30: Q1:n;k:k+1 := Q1:n;k:k+1 ;
−s c
31: end for/* Implicit QR step ends here */
32: if |bm | < ε(|am−1 | + |am |) then /* Check for convergence */
33: m := m − 1;
34: end if
35: end while
4.7. RESEARCH 87
Algorithm 4.6 shows the implicit symmetric tridiagonal QR algorithm. The shifts
are chosen acording to Wilkinson. An issue not treated in this algorithm is deflation.
Deflation is of big practical importance. Let us consider the following 6 × 6 situation
 
a1 b2
 b2 a2 b3 
 
 b3 a3 0 
T =

.

 0 a 4 b5 
 b5 a 5 b6 
b6 a 6

The shift for the next step is determined from elements a5 , a6 , and b6 . According to (4.12)
the first plane rotation is determined from the shift and the elements a1 and b1 . The im-
plicit shift algorithm then chases the bulge down the diagonal. In this particular situation,
the procedure finishes already in row/column 4 because b4 = 0. Thus the shift which is an
approximation to an eigenvalue of the second block (rows 4 to 6) is applied to the wrong
first block (rows 1 to 3). Clearly, this shift does not improve convergence.
If the QR algorithm is applied in its explicit form, then still the first block is not treated
properly, i.e. with a (probably) wrong shift, but at least the second block is diagonalized
rapidly.
Deflation is done as indicated in Algorithm 4.6:

if |bk | < ε(|ak−1 | + |ak |) then deflate.

Deflation is particularly simple in the symetric case since it just means that a tridiagonal
eigenvalue problem decouples in two (or more) smaller tridiagonal eigenvalue problems.
Notice, however, that the eigenvectors are still n elements long.

4.7 Research
Still today the QR algorithm computes the Schur form of a matrix and is by far the
most popular approach for solving dense nonsymmetric eigenvalue problems. Multishift
and aggressive early deflation techniques have led to significantly more efficient sequential
implementations of the QR algorithm during the last decade. For a brief survey and a
discussion of the parallelization of the QR algorithm, see [7].
The three steps of the presented symmetric QR algorithm are (1) reducion of the origi-
nal matrix to tridiagonal form, (2) computation of the eigenpairs of the tridiagonal matrix,
and (3) back-transformation of the eigenvectors. In the ELPA project the first step has
been successfully replaced by a two-stage procedure: transformation full to banded, and
banded to tridiagonal. This approach improves the utilization of memory hierarchies [8, 3].

4.8 Summary
The QR algorithm is a very powerful algorithm to stably compute the eigenvalues and (if
needed) the corresponding eigenvectors or Schur vectors. All steps of the algorithm cost
O(n3 ) floating point operations, see Table 4.1. The one exception is the case where only
eigenvalues are desired of a symmetric tridiagonal matrix. The linear algebra software
package LAPACK [1] contains subroutines for all possible ways the QR algorithm may be
employed.
88 CHAPTER 4. THE QR ALGORITHM
nonsymmetric case symmetric case
without with without with
Schurvectors eigenvectors
10 3 14 3 4 3 8 3
transformation to Hessenberg/tridiagonal form 3 n 3 n 3n 3n
20 3 50 3
real double step Hessenberg/tridiagonal QR al- 3 n 3 n 24n2 6n3
gorithm (2 steps per eigenvalues assumed)
4 3
total 10n3 25n3 3n 9n3

Table 4.1: Complexity in flops to compute eigenvalues and eigenvectors/Schur vectors of

a real n × n matrix

We finish by repeating, that the QR algorithm is a method for dense matrix problems.
The reduction of a sparse matrix to tridiagonal or Hessenberg form produces fill-in, thus
destroying the sparsity structure which one almost always tries to preserve.

Bibliography
[1] E. Anderson, Z. Bai, C. Bischof, J. Demmel, J. Dongarra, J. D.
Croz, A. Greenbaum, S. Hammarling, A. McKenney, S. Ostrouchov,
and D. Sorensen, LAPACK Users’ Guide – Release 2.0, SIAM, Philadel-
phia, PA, 1994. (Software and guide are available from Netlib at URL
http://www.netlib.org/lapack/).

[2] P. Arbenz and G. H. Golub, Matrix shapes invariant under the symmetric QR
algorithm, Numer. Linear Algebra Appl., 2 (1995), pp. 87–93.

[3] T. Auckenthaler, H.-J. Bungartz, T. Huckle, L. Krämer, B. Lang, and

P. Willems, Developing algorithms and software for the parallel solution of the sym-
metric eigenvalue problem, J. Comput. Sci., 2 (2011), pp. 272–278.

[4] J. W. Demmel, Applied Numerical Linear Algebra, SIAM, Philadelphia, PA, 1997.

[5] J. G. F. Francis, The QR transformation – Parts 1 and 2, Comput. J., 4 (1961-

1962), pp. 265–271 and 332–345.

[6] G. H. Golub and C. F. van Loan, Matrix Computations, The Johns Hopkins
University Press, Baltimore, MD, 2nd ed., 1989.

[7] B. Kågström, D. Kressner, and M. Shao, On aggressive early deflation in

parallel variants of the QR algorithm, in Applied Parallel and Scientific Computing
(PARA 2010), K. Jónasson, ed., Heidelberg, 2012, Springer, pp. 1–10. (Lecture Notes
in Computer Science, 7133).

[8] A. Marek, V. Blum, R. Johanni, V. Havu, B. Lang, T. Auckenthaler,

A. Heinecke, H.-J. Bungartz, and H. Lederer, The ELPA library: scalable
parallel eigenvalue solutions for electronic structure theory and computational science,
J. Phys.: Condens. Matter, 26 (2014), p. 213201.

[9] B. N. Parlett, The QR algorithm, Computing Sci. Eng., 2 (2000), pp. 38–42.
BIBLIOGRAPHY 89
[10] H. Rutishauser, Solution of eigenvalue problems with the LR-transformation, NBS
Appl. Math. Series, 49 (1958), pp. 47–81.

[11] J. H. Wilkinson, The Algebraic Eigenvalue Problem, Clarendon Press, Oxford, 1965.
90 CHAPTER 4. THE QR ALGORITHM