If $y:[0,\infty)\to[0,\infty)$ is a continuously differentiable function satisfying $$y(t)=y_0-\int\limits_0^t y(s)ds$$ for $t\ge0$, then
- $y^2(t)=y_0^2+\left(\int\limits_0^t y(s)ds\right)^2-2y_0\int\limits_0^t y(s)ds$
- $y^2(t)=y_0^2-2\int\limits_0^t y^2(s)ds$ (CSIR December 2014)
(My attempt has been converted into an answer since no one contradicted so far, see below.)
Trivially, 1 is true.
For 2, using Leibniz integral rule:
$\dot y(t)=-y(t)$
$\dot y^2(t)=2y(t)\dot y(t)=-2y^2(t)$
Integrating from 0 to t,
$$\int\limits_0^t d(y^2(s))=y^2(t)-y^2(0)=-2\int\limits_0^t y^2(s)ds$$
And hence 2.