Probability and Statistics Masterclass - Making maths fun.

Understand what a random variable is through clear examples of discrete and continuous types. See how constants and independent variables interact in a random experiment to produce outcomes.

Explore nominal and ordinal data within the broader eight data types, distinguishing categorical and numerical forms, and clarifying qualitative versus quantitative interpretations using practical examples.

Explore central tendency concepts such as mean, median, and mode, how outliers skew the mean, why median suits ordinal data, and how bimodal and multimodal data arise.

Explore central tendency through practical examples, showing that mode works for categorical data like states or ratings, while mean and median depend on data being numerical or ordered.

Explore data visualization by distinguishing categorical and numerical data, using bar charts and pie charts for categories, and histograms for numerical data, with notes on mean, median, and mode.

Discover how percentiles, quartiles and the interquartile range describe data distribution, with the median as the 50th percentile, and learn percentile calculations and the concept of maximum.

Explore how to compute mean, median, and mode, along with range, percentiles, and interquartile range, then read a box plot and understand how skewness affects these measures.

Learn how standard deviation and variance quantify data spread around the mean, and distinguish population and sample formulas. Use the coefficient of variation to compare variability across datasets.

Understand why sample standard deviation uses n minus one in the denominator rather than n, to correct bias when estimating from the sample mean.

Covariance measures how two variables move together, indicating positive, negative, or no relation. It links to variance and leads to correlation, a unitless value between -1 and 1.

Explore the normal distribution and its use in chi square tests for association of categorical variables, noting that the area under the curve equals one.

Explore the chi square distribution, defined as the sum of squares of k standard normal variables with k degrees of freedom, and read tables to assess association between categorical variables.

Explore the chi-square goodness-of-fit test, defining null and alternative hypotheses, computing the chi-square statistic from observed and expected counts, and assessing fit with degrees of freedom and 5% significance level.

Explore how the chi square distribution tests association between two categorical variables by comparing observed and expected counts, using a real airline example to reject or not reject the null.

Learn to measure the linear association between two numerical variables with the Pearson correlation, computed from covariance divided by standard deviations, and use Spearman for monotonic, non linear trends.

Explore how the probability distribution function represents each value's probability as a graph, with the discrete case using a probability mass function and a fair die example.

Explore continuous random variables and the distribution function, and the probability density function, where the area under the curve between two heights yields probability and the total area equals one.

Explore the Bernoulli distribution, a discrete model for two-outcome experiments with probability p and 1-p, including coin toss examples and Bernoulli trials.

Explore the binomial distribution as an extension of the Bernoulli model, with independent trials and the binomial formula for exactly k successes in n trials, illustrated by coin flips.

Understand the expected value as the weighted average of outcomes. A gambling example shows how positive value supports playing and relates to the law of large numbers.

Calculate the expected value of changes in sales from 50 days with outcomes 2000, 0, and -1000, based on their probability distribution, illustrating the long-term average and standard deviation.

Learn to compute the expected value for a Bernoulli distribution by weighting outcomes 0 and 1 with their probabilities, showing that the expected value equals p, the probability of success.

Explore the expected value of the binomial distribution using n p, illustrated with six balls per over and p = 0.3; 1.8 is the long-run average.

Explore how the law of large numbers makes the observed average converge to the expected value as trials grow, using dice and overs to show increasing confidence in results.

Explore the normal distribution, the bell curve with a peak at the mean; area under the curve yields probabilities between values, and mu and sigma govern 68, 95 and 99.7.

Show how standard deviation and variance shape the normal distribution: higher standard deviation flattens the curve, lower standard deviation peaks it, while the area under the curve remains one.

Explore the cumulative distribution function by recapping histograms and pdfs, showing how the CDF encodes area under the pdf up to a value, and contrasting it with cumulative histograms.

Explore the normal distribution formula, identify its parameters mu and standard deviation, and see how changing these values shapes the curve, with hands-on Excel data experiments.

Explore the normal distribution using an Excel utility to compute pdf and cdf, adjust mu and sigma, and visualize how mean and spread shape the area under the curve.

Learn how to convert any normal distribution to the standard normal (unit normal) with mean zero and unit variance using the Z-score transformation, enabling use of a single Z table.

Explore the normal distribution with an intuitive view of extreme values. See how the asymptotic curve assigns positive, though tiny, probability to all numbers from minus infinity to plus infinity.

Learn how the z score standardizes any normal variable by subtracting the mean and dividing by the standard deviation, converting to a unit normal distribution.

Explain the z score in detail, showing how x minus mu divided by sigma measures how many standard deviations a value lies from the mean in the unit normal distribution.

Learn to read a z score table for a normal distribution, using the z formula (x - μ)/σ, interpreting left and right areas and symmetry to find probabilities.

Calculate the probability of a free pizza using the z score for a normal distribution, with mean 16.3 and standard deviation 0.2, finding the left-tail area below 16 inches.

Compute the probability of a pizza size above 16.5 using z-scores in a normal distribution with mean 16.3 and sd 0.2, and apply symmetry to read left-tail areas.

Apply z scores to a normal distribution with mean 70 and standard deviation 5 to determine how many students score below 65, above 75, and between 65 and 75.

Use z-scores to infer mu and sigma from 15% below 30 hours and 10% above 50 hours. Solve two linear equations from these z-values and confirm the normal distribution properties.

Study symmetric distributions and skewness to visualize distributions without data. See how the normal distribution centers mean, mode, and median, and how skewness indicates asymmetry.

Explore the central limit theorem: sample means from any distribution become normal around the population mean with standard deviation divided by sqrt(n) for samples of size at least 30.

The central limit theorem shows that with samples over 30, sample means become normal with mean equal to population mean and sd equal to population sd divided by sqrt(n).

The central limit theorem states that the distribution of sample means is normal with mean mu and standard deviation sigma over sqrt(n); for population, n must be at least 30.

Combine the central limit theorem with z scores to solve sample mean problems. Model the sample mean as normal, compute z values, and read areas from the z table.

Apply the central limit theorem to determine the probability that the sample mean of 49 hours lies between 441 and 446, with mu 448 and standard error 3, about 24.5%.

Examine discrete and continuous uniform distributions, where discrete assigns equal probability to finite values (like a fair die), and continuous uses a constant density on [A, B].

Explore the log normal distribution and its link to the normal distribution via log and exponentials, and note mu and sigma from the normal curve of log X.

Explore how lognormal distributions appear across real-world contexts—from internet comments and dwell times to chess games and medical sizes—highlighting its ubiquity beyond normal distributions.

Explore power law distribution, where y grows as a power of x and exhibits exponential rise or fall; emphasize positive x and a positive coefficient for valid probability distributions.

Explore the pareto distribution as a power law with alpha, illustrating wealth and income allocation through the 80/20 rule and real-world examples.

Explore the Pareto distribution through its formula, compare it to a power law, and examine how alpha (shape) and beta (minimum value) shape the distribution and its graph.

Learn how a Q-Q plot compares an unknown distribution to a known one by pairing ordered quantiles; a straight line indicates they are the same distribution, widely used beyond normality.

Explore the box-cox transformation that converts any distribution toward normality using the lambda parameter, with lambda=0 yielding a log transform. Visualize the results with QQ plots to assess effectiveness.

Explore how distributions guide real-life decisions, from lognormal and normal sizing to Pareto 80/20 storage planning and hypothesis testing for a data analyst, machine learning engineer, or data scientist.

Explore hypothesis testing, linking normal distribution and central limit theorem to null and alternate hypotheses, with alpha and P value guiding evidence.

Construct a 90% confidence interval for mean number of years US companies trade with firms in India, using a 44-sample mean 10.455 and sigma 7.7, per the central limit theorem.

Compute a 98% two-tailed confidence interval for the average age using the sample mean 34.3 years, n = 50, and population standard deviation eight years, yielding 31.67 to 36.93 years.

Learn why the z score fails when population std dev is unknown and how the t score, using sample std dev, degrees of freedom, and the t table, replaces it.

Test the null mu=20 against mu<20 using a one-tailed t-test with n=20, x̄=19.8, s=3.1, applying the central limit theorem and interpret p-values to conclude the null cannot be rejected.

Assess a one-tailed hypothesis test for plant growth, using sample mean 11.4, sd 2.5, n=15 to compute t and p value and alpha-driven rejection region.

Demonstrates a one-tailed t-test with n=25, mean 85, sd 4.1 against mu=82 at 0.05, showing 3.65 > 1.71 and rejecting the null, with correct hypothesis setup.

Reject the null hypothesis when the p-value is less than alpha; never say we accept it, and interpret evidence as weak, moderate, or strong based on alpha and p-values.

Explore the fundamentals of probability through common examples like coin tosses and dice, and learn how long-term relative frequency defines probability, with concepts of sample space and events.

Explore the laws of probability, including the sum of all outcomes equals one, probe conditional probability and Bayes' theorem, and study sample space through coin and ball examples.

Examine mutually exclusive and independent events with coin tosses and samples, simplify probability with intersection and union, and apply conditional probability and Bayes theorem to solve problems.

Practice probability with intuitive reasoning and simple formulas, using union and intersection concepts while exploring how permutation and combination relate to basic formulas through multiple examples.

Compute the probability of drawing two blue and two green balls from four blue and three green without replacement, using intuitive sample-space reasoning and mutually exclusive orderings to obtain 18/35.

Explore the chain rule of probability and its roots in conditional probability and Bayes theorem. See how independence and mutual exclusivity simplify A intersection B and P(A|B) notation.

Explore the chain rule of probability using a 52-card deal to four players, illustrating conditional probability and how to compute the chance each player receives one ace.

Explore conditional probability through a chain rule card example, showing how to compute the probability that four players each receive one ace from a 52-card deck.

Explore the birthday paradox by calculating the probability that ten people have all distinct birthdays among 365 days and its complement of shared birthdays using independence and conditional probability.

Explore probability through a three-friend chit problem and a five-day power-cut scenario, analyzing exactly one or none picks their name and at least one power cut using independence.

Explore real-world probability reasoning through cricket scenarios and independence of components, comparing tie-and-win strategies with hitting four or six, and calculating the probability that all ten components work.

Explore probability through an eight-story elevator scenario and a five-question quiz, analyzing independence, 70% success, and strategies for solving at least two or all five correctly.

Explore probability trees to visualize and solve complex problems with diagrams, turning scenarios into clear branches and using conditional probabilities to compute events like watching cricket and seeing ads.

Learn the total law of probability, built on conditional probability, by partitioning outcomes into mutually exclusive cases and applying it to complements and multiple events.

Compute the probability of a tape recorder having a flaw given it died, using conditional probability and the total probability law, by translating statements into math and handling missing information.

Explore how to evaluate defect probabilities in a ten-item sample, showing why each item has a 10% defect rate, and apply mutual exclusivity and conditional probability.

Apply the total law of probability to a travel luggage problem, using conditional probabilities and probability trees to find the chance of luggage arriving in B given on-time or late flights.

Learn how counting underpins probability, distinguish permutations from combinations, and see when order matters or not, using real examples and the idea of numerator and denominator.

Count outcomes to teach probability with mutually exclusive travel options from Delhi to Mumbai, compute total ways and train probability, and introduce permutation and combination as counting tools.

Multiply options for legs—Delhi to Mumbai: 8 flights, 12 trains, 2 roadways; Mumbai to Chennai: 6 flights, 10 trains, 2 roadways. Treat the legs as independent and yield 396 ways.

Explore how constraints create dependencies between travel legs in permutation and combination problems, and learn how mutually exclusive, independent, and dependent events affect the number of possible ways.

Learn to count using combinations and independence: select one representative from physics, chemistry, and mathematics, and count five-digit ternary sequences with first-digit constraints.

Explore counting six-digit natural numbers with specific odd and even digit constraints, treating zero as even, allowing repetition, and using permutation and combinations ideas with independence and multiplication.

Explore counting distinct terms in products of sums using permutation and combinations, with four by five terms yielding 20 ways and a general M, N, T terms case.

Explore factorial notation, define n factorial as the product 1 through n, show examples like 5! and 3!, and explain how n! equals n times (n minus 1)!, enabling cross-cancellation.

Define permutation as an ordered arrangement of k distinct elements from n and learn the nPk formula, including edge cases and when listing scenarios matters.

Explore permutation with repetition where order matters and repeats are allowed, using n distinct items and k selections. See why the formula becomes n^k.

Explore permutation scenarios where N is distinct and K may repeat, including when K equals N. Learn when to apply npq and npk formulas, and why dividing by repeating-factorial matters.

Explore permutation with repetition and constraints by counting arrangements of words like probability and Mississippi using factorials and division by repeated letters.

Explore permutation with repetition by counting four-digit arrangements from a repeating set, using nPk divided by repetition factorial, and recognizing when the formula fails, with worked examples.

Explore when permutation formulas with repetition hold or fail, analyze scenarios like three as and three bs, and verify arrangements through intuitive case analysis.

Learn permutation strategies by treating vowels as a single unit, handle repeats like two b's, and analyze scenarios where consonants are grouped as well.

Apply permutation principles to words like correction and collection, counting vowel-together and not-together cases, handling duplicates like two r's and two c's, using seven positions with factorials to compute totals.

Examine permutations with repetition using A, B, C. Positions yield 9 arrangements; repeating characters don't affect count when repetition is allowed, with three and four digits giving 27 and 81.

This lecture introduces combinations, contrasts them with permutations, and demonstrates unordered selections from n distinct elements using the nCk formula, with committee examples including exact and at least two women.

Learn to distinguish permutations from combinations using practical examples, such as top three gifts with different amounts versus equal amounts, and committees where order of responsibilities matters.

Explore circular table permutation problems with four men and four women alternating seats; fix one person, apply 4P4 for men and 3P3 for women, yielding 144 arrangements.

Treat this as a combination problem: choose two fiction from n and two nonfiction from six to total 150 ways. Thus C(n,2) times C(6,2) = 150, so n = 20.

Defines expected value as the weighted average of outcomes for a random variable, uses a gambling example, and shows profitability over many trials, illustrating the law of large numbers.

Compute the expected value from a discrete distribution and relate it to the mean and variance; apply x minus mu squared with probabilities to find standard deviation.

Explore the first property of expected value: adding a constant to a random variable shifts its mean by that constant, illustrating E[X+C] = E[X] + C with simple examples.

Explore how adding a constant to x does not change the variance or standard deviation, because both x and the mean, the expected value, shift by the same amount.

The lecture states E[CX] = C E[X], Var(CX) = C^2 Var(X), and that standard deviation scales by |C| when multiplying X by C.

Explore the properties of expected value and variance for linear transformations and sums, including independence conditions, with examples to reinforce how A and B affect E[X], Var[X], and sum X+Y.

Compare hosting a cricket match in England or India by calculating expected value, variance, standard deviation, and the coefficient of variation to judge the more profitable, risk-adjusted option.

Evaluate two-ball draws from a bag of eight white, three black, two red; assign +10 for black, -2 for white, 0 for red; determine expected value, probabilities, and three-play profitability.

Learn to calculate the probability of at least one power cut over the next three days with a daily 0.05 chance, by summing scenarios or using 1 minus no cuts.

Examine a two-child probability puzzle about a girl's waiver, compare 1/3 and 1/2 solutions, and conclude there is a 50% chance the neighbor has two girl children.

Apply conditional and total probability to a four-choice question to determine P(knows the answer | correct) = 6/7.