Up to this point you have worked with continuous distributions mostly as formulas: a density function, an integral, a z-score table. This topic shifts the emphasis to judgement. Given a real data set or a described engineering situation — machining tolerances, time between equipment failures, arrival times, repair costs — which continuous model is defensible, and what does the resulting probability actually tell a decision-maker?
The practical skill has three parts. First, matching the physical behaviour of the process to the shape and assumptions of a candidate distribution: constant likelihood across an interval suggests uniform, memoryless waiting between independent events suggests exponential, the accumulation of many small independent effects suggests normal, and multiplicative effects often produce a right-skewed lognormal. Second, computing the probability correctly. Third, checking whether the assumption held at all, and reporting the result honestly, including its limitations.
As you watch the videos below, pay attention to the reasoning that precedes each calculation: what feature of the scenario justified the choice of model? Note also where the presenters mention related distributions — the gamma and lognormal are not required computations in this module, but seeing them helps you understand why the uniform, normal, and exponential are chosen for some problems and not others.
A short, worked introduction to the continuous uniform distribution, where every outcome in an interval is equally likely. Based on its title and description, the video should set up the density and cumulative forms and then solve example probability problems step by step.
It gives direct, example-driven coverage of the uniform distribution subtopic at a level suited to a first course, and its focus on solving problems supports the calculation and interpretation outcomes.
Steve Brunton presents the gamma distribution and explains its relationship to the exponential distribution — specifically, as the waiting time until the r-th event in a Poisson process. Expect discussion of how the exponential arises as a special case and how the shape changes with its parameters.
Seeing the exponential placed inside a larger family clarifies exactly what the exponential assumes about waiting times, which is central to selecting and validating it in professional contexts; Brunton is a respected engineering faculty producer known for clear, correct explanations.
codebasics introduces the lognormal distribution as a right-skewed model and points to real-world examples such as income and length of hospital stay. The video is likely to contrast it with the symmetric normal distribution and show how a log transformation relates the two.
It directly addresses the risk of assuming normality when data are skewed, reinforcing the distribution-fit and assumption-validation subtopics and sharpening your sense of when the normal model is and is not appropriate.
Together these videos move beyond formula practice toward model selection and interpretation. The uniform video supplies direct computational practice with a named continuous distribution and its density and cumulative forms (MO1, MO2). The gamma video clarifies the exponential's role as a waiting-time model within a Poisson process, giving you a principled basis for choosing it over other candidates in engineering reliability and queueing settings (MO3). The lognormal video provides a concrete counterexample to automatic normality assumptions, prompting you to check data shape before modelling and to qualify the conclusions you draw (MO3, MO4). Across all three, the recurring emphasis on real examples supports interpreting computed probabilities as evidence for decisions rather than as answers in isolation.
These videos were chosen from YouTube on October 2, 2026. Each has captions; the player turns them on. Videos on YouTube can be changed or removed — if one no longer plays, tell your instructor. If a player does not appear, use its Watch on YouTube link.