An integrating factor for a differential one-form is a nonzero function for which is an exact differential, say . Then solutions lie on level sets . Locally, the compatibility condition is .
Here multiplication by , which is nonzero for , produces
The initial condition makes this first integral equal to . Taking the positive branch yields the solution
on the connected interval containing zero where the radicand is positive. This domain restriction matters because the original differential equation is singular at .
Completing the square gives . As , the differential equation consequently implies
At , . Thus, moving a little to the left from , the curve rises above . Throughout every segment of the direction field points downwards as increases, by the inequality above. Tracing backwards therefore makes rise further, so it cannot return to . This graphical barrier argument proves and for every on the solution. The explicit first integral also shows this branch extends throughout : writing , the minimum of is , so for . Including the initial point, for all .
For the curve is on , where the strictly increasing function has the unique positive root at . The curve decreases through with slope and reaches only as a limiting endpoint. Its derivative is
As , and . As , the numerator tends to and the denominator to zero from above, so again . More precisely, at the endpoint. Both ends have the requested unbounded negative slope.
Figure 1.
The positive solution for initial value one, with the horizontal barrier and the steep limiting endpoint
. Details.
After independent tosses, an even count arises in exactly two disjoint ways: an even count after tosses followed by a tail, or an odd count followed by a head. Independence of the last toss from the preceding ones therefore gives
Subtracting the fixed point turns this into the linear recurrence
As zero is even, , so the probability is
For this is for every ; for it is always one; for it alternates between one and zero according to the parity of .
The indicator function equals one when the event occurs and zero otherwise. Its expected value is . By linearity of expectation, no independence assumption is needed to obtain
Since , including , the variance is
In particular, correlations between the events cannot simply be discarded.
Write . If , then implies . Applying Chebyshev's inequality therefore gives
Equivalently, the contribution of to is . A positive mean is necessary for the displayed ratio to be defined. If , nonnegativity instead gives almost surely, and the printed ratio would be .
This is a separable differential equation. On the solution with , integration from the initial condition gives
The right side lies in , so taking its sine stays on the required branch. The solution is
Near the origin, , so and the tangent line is horizontal. At the other end, , while
Thus the curve has a vertical limiting tangent line as , even though is excluded from the domain.
For the direction field, the slope is nonnegative everywhere in the square. The segments are horizontal on and on , become steeper as increases at fixed , and become flatter as increases at fixed . Near they are almost vertical unless is correspondingly close to one. The excluded corner has no single limiting slope: the ratio depends on the path of approach.
Figure 1.
Positive slope field and the solution from the origin, with its horizontal and vertical limiting tangents
. Details.
The characteristic equation of the homogeneous linear differential equation is , giving the homogeneous solution . For the first forcing, a particular solution of the form gives , so
The initial conditions impose and , whence , . Thus the first solution is
For the second forcing, is already a homogeneous solution, so the usual exponential trial must be multiplied by . Substitution of the particular solution gives , hence . Now , and the initial conditions give , . Therefore the resonant solution is
The additional factor reflects resonance in a differential equation.

There are unlisted articles, also show them or only show them.