Continuous Random Variables and Density Curves

With a discrete random variable you can list outcomes and give each one a probability, like a bar in a histogram with a specific height. A continuous random variable, like height or time, can take any value in a range, so there are infinitely many possible outcomes packed into any interval. You can't assign a positive probability to each one without the total blowing past 1. The fix: describe the variable with a density curve, a smooth curve where probability is area underneath it over an interval, not the height of a bar at a point.

This is why asking for the exact probability a person is 70.000... inches tall gives zero. That single height is a point on the horizontal axis with no width at all, and area needs width. Zero width times any height still gives zero area, so zero probability. This feels strange at first because the person obviously has some height, but the density curve only assigns probability to ranges, like between 69.5 and 70.5 inches, never to a single infinitely precise value.

The simplest density curve is the uniform distribution, a flat rectangle where every equal-length interval has equal probability. Because the whole area under any density curve must equal 1, you find probabilities for a uniform variable just by comparing interval lengths, which sets up the reasoning you'll reuse for the normal curve next, where area is found differently but the same logic of area-equals-probability still holds.