Stochastic differential equations · Part 7 of 10
When diffusions explode: Feller's test
The existence theorem for stochastic differential equations assumes coefficients that grow at most linearly. Many natural models violate that assumption, and their solutions can run off to infinity in finite time. In one dimension there is a complete answer to whether this happens, due to William Feller.
Solutions up to explosion
Suppose and are locally Lipschitz on an interval , whose endpoints may be infinite, and that on . Applying the existence theorem on larger and larger closed subintervals gives a unique solution up to the explosion time
where is the first time the solution leaves the -th subinterval. The solution explodes if , which means reaches an endpoint of in finite time. When , this is a genuine blow-up to or .
Two key functions
Fix a reference point . The scale function is
It is built so that , which makes a local martingale: in the coordinate , the process has no drift. Hitting probabilities are ratios of differences of the scale function, as the interval example in the previous post showed for Brownian motion with drift.
Scale describes where the process goes, but not how fast. Timing enters through the speed measure , and the two combine into the function
This function is nonnegative on both sides of , and it measures how long the process takes to travel toward each endpoint.
Feller’s test
Theorem. The solution never explodes, that is for every starting point in , if and only if
If is finite at an endpoint, explosion through that endpoint has positive probability, and the scale function decides how likely it is. For example, if and , the process drifts toward , and it explodes with probability one exactly when .
The intuition is that a finite means the expected time to travel all the way to is finite. A process can be transient, drifting toward an endpoint forever, without ever reaching it in finite time.
Examples
Brownian motion. With and constant , the scale function is and , which is infinite at both ends. There is no explosion, as expected.
Brownian motion with drift. With constants and , write . Then is bounded above, so the process drifts to , but , so it never gets there in finite time. This is transience without explosion.
A superlinear drift, with and without noise. The deterministic equation blows up. Adding noise does not help. For , the scale density is , which is integrable at but not at , so the process is pushed toward . For large , the inner integral in behaves like , so the integrand of decays like and . The process explodes with probability one. Noise can prevent explosion only when it is strong enough to compete with the drift, for example when also grows quickly, and Feller’s test quantifies exactly how strong it needs to be.
Reaching a boundary at zero. The same functions decide whether a process on can hit zero. Geometric Brownian motion never does. The squared Bessel process hits zero when its dimension is below 2 and never hits it when . The threshold matches the fact that a two-dimensional Brownian motion never hits a given point.
Why it matters
Explosion is not just a pathology. Models in finance, population biology and physics often have superlinear coefficients, and Feller’s test tells you whether such a model makes sense for all time. It is also a good example of a general theme: in one dimension, the scale function and speed measure determine essentially everything about a diffusion, which is why one-dimensional diffusions are so much better understood than higher-dimensional ones.