If I take $n$ linearly independent vectors ($n \times 1$), $v_1, v_2, \ldots, v_n$ and construct a matrix which is the sum of their outer products,

$$M = \sum_i^n v_iv_i^t$$

Then, can it be proven that $M$ is always full rank?


closed as off-topic by Namaste, José Carlos Santos, user91500, John B, J. M. is a poor mathematician Aug 28 '17 at 5:15

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – Namaste, José Carlos Santos, user91500, John B, J. M. is a poor mathematician
If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    $\begingroup$ Such a matrix will be positive definite. $\endgroup$ – achille hui Aug 25 '17 at 22:57
  • $\begingroup$ Can we prove this? $\endgroup$ – Rohit Pandey Aug 25 '17 at 23:01
  • $\begingroup$ ^Exploit the fact that the $\{v_i\}$ form a basis. $\endgroup$ – stochasticboy321 Aug 25 '17 at 23:15

If $M = \sum v_i v_{i}^{T}$ then

\begin{align} Mx & = \sum v_i v_i^T x \\[10pt] & = \sum v_i (v_i^T x) \\[10pt] & = \sum (c_i)v_i \end{align}

where $c_i$ is a scalar denoting the inner product of $x$ and $v_i$. Since the $v_i's$ are linearly independent this sum yields the zero vector if and only if $c_i=0$ for all $i$ and hence $x$ is the zero vector. So your matrix clearly has full rank.


First, notice that $M$ being full-rank is equivalent to the induced linear operator being injective, which is equivalent to $\det M \neq 0$.

Now define $T : e_i \mapsto v_i$, where $e_i$ is the canonical basis of $\mathbb R^n$. Then $v_i^T = (T e_i)^T = e_i^T T^T$, hence $$M = \sum_i T e_i e_i^T T^T = T \left( \sum_i e_i e_i^T \right) T^T$$ But the matrix in brackets is just the identity matrix, thus $$M = T T^T \quad \Longrightarrow \quad \det M = ( \det T)^2 $$ But $T$ is full rank, thus $\det T \neq 0$ and $\det M \neq 0$.

  • $\begingroup$ But the existence of such a $T$ is contingent on $M$ being invertible, no? $\endgroup$ – Rohit Pandey Aug 25 '17 at 23:16

Not the answer you're looking for? Browse other questions tagged or ask your own question.