A probability density function, or PDF, describes how probability is spread across the possible values of a continuous random variable. Instead of assigning probability to one exact value, it assigns density along the number line. This matters because measurements like height, time, mass, and voltage are often modeled as continuous quantities.
Calculus connects the curve to probability through area under the graph.
Understanding Calculus: Probability Density Functions
A density is not the same thing as a direct chance. Its size must be interpreted together with a small width on the horizontal axis. A narrow region can have a high density but still contain little probability.
A wide region with a lower density can contain more probability overall. This is why a density curve may rise above one without breaking any rule. The vertical units matter too.
If a variable is measured in seconds, the density has units of per second. Multiplying density by a width in seconds leaves a unitless probability. This unit check helps students see why areas, rather than graph heights alone, are meaningful.
A useful partner to a density function is the cumulative distribution function. It records the probability accumulated from the far left up to a chosen value. As the chosen value moves right, this accumulated amount can only stay the same or increase.
Calculus links the two graphs closely. The cumulative function is built by integrating the density, while the density is the rate at which cumulative probability is increasing.
The probability between two readings can then be found by subtracting the earlier cumulative value from the later one. This method is especially useful when software provides a cumulative table instead of a formula for the density.
The expected value gives a long run average, but it does not have to be a value that is likely to occur. It is a balance point created by weighting each possible value according to its density. Values far from the center can influence this balance strongly if the curve has long tails.
For this reason, a mean alone does not fully describe a distribution. Spread is commonly measured using variance, which weights the squared distance from the mean.
Squaring makes negative and positive distances contribute equally, while giving extra importance to large deviations. In real settings, this matters for manufacturing tolerances, travel times, test scores, and measurement errors.
Students should pay close attention to the range where the model applies. A formula may describe density only between two endpoints, with no possible values outside them. Integrating beyond that range adds nothing, but forgetting the boundary can lead to an incorrect answer.
Another common mistake is treating a graph as equally likely wherever it exists. Curves can be symmetric, skewed, flat in places, or made from separate pieces. When reading a graph, first identify the interval of interest, then compare areas over that interval.
For complicated curves, numerical methods approximate the needed area with many thin rectangles. More and thinner rectangles usually give a better estimate.
Key Facts
- For a valid PDF, f(x) >= 0 for all x.
- The total area under a PDF is 1: integral from -infinity to infinity of f(x) dx = 1.
- Probability over an interval is area: P(a <= X <= b) = integral from a to b of f(x) dx.
- For a continuous random variable, P(X = a) = 0 because a single point has zero width.
- The mean or expected value is E[X] = integral from -infinity to infinity of x f(x) dx.
- If f(x) = 0 outside an interval, only integrate over the interval where the density is nonzero.
Vocabulary
- Probability density function
- A function f(x) that gives probability density for a continuous random variable and whose total area is 1.
- Continuous random variable
- A variable that can take any value in an interval rather than only separated values.
- Area under the curve
- The integral of a density function over an interval, which represents probability.
- Expected value
- The long run average value of a continuous random variable, found by integrating x f(x).
- Normalization
- The requirement that the total integral of a probability density function equals 1.
Common Mistakes to Avoid
- Treating f(a) as P(X = a), which is wrong because density is not the same as probability. For a continuous variable, the probability at one exact point is 0.
- Forgetting that total area must equal 1, which makes the function invalid as a probability density function. Always check that the integral over the whole domain is 1.
- Using height instead of area to compare probabilities, which is wrong because interval width also matters. Probability is found by integrating over an interval.
- Integrating over the wrong bounds, which gives the probability of the wrong event. Match the lower and upper limits to the exact interval stated in the problem.
Practice Questions
- 1 Let f(x) = 2x for 0 <= x <= 1 and f(x) = 0 otherwise. Verify that f(x) is a valid probability density function by computing the total integral.
- 2 For f(x) = 2x on 0 <= x <= 1, find P(0.25 <= X <= 0.75).
- 3 A probability density curve is tall and narrow near x = 2 but shorter and spread out from x = 5 to x = 9. Explain why the wider region might represent a larger probability even though the curve is lower there.