Uniform Product > 1/2 Chance

This question considers a pair of independent continuous random variables with uniform distributions over a unit square, and asks for the probability that their product exceeds a certain threshold. The setup turns a probabilistic event into a geometric region in the plane, defined by a curve inside the square. Because of the symmetry and simplicity of the uniform distributions, the event can be visualized directly as a subset of the square, and the probability becomes the area of that subset relative to the whole. Variants of this question often appear in quantitative interviews to see whether candidates can translate between algebraic conditions on random variables and geometric pictures.

Solving it leans on interpreting joint densities, basic multivariable integration, and comfort with inequalities that define regions in the plane. Candidates need to recognize independence and uniformity as cues that the joint density is constant, justify the geometric approach, and correctly determine where the boundary curve intersects the edges of the domain. Interviewers watch for clear reasoning in choosing integration limits, correct handling of continuous probabilities, and the ability to explain the connection between analytic integration and the area picture.

What it tests

When two independent continuous random variables are uniformly distributed over a rectangle, the probability that a function of them exceeds a threshold is the area of the region in the rectangle where the condition holds, divided by the total area. This reduces probability questions to geometric area calculations. The boundaries of the region are determined by the equation defining the event, and the limits of integration are set by the intersection of this boundary with the edges of the rectangle. The key is that independence and uniformity make the joint density constant, so probability is proportional to area. This geometric approach generalizes to any event that can be described as a region in the plane (or higher dimensions), not just products or hyperbolas.