Complete Probability & Statistics for Data Science

Explore percentile and its link to quartiles, learn to rank observations and identify outliers, compute percentile rank with 100×(values before x)/n, and estimate percentile values by averaging surrounding observations.

Learn to identify and remove outliers using the five number summary, compute q1, q3, and the iqr, and plot a box plot with min, q1, median, q3, and max.

Explore the normal distribution, its bell-shaped symmetry, and the empirical 68–95–99.7 rule, then learn z-scores, standardization, and how to use the z table for data analysis.

Identify and remove outliers in data sets using Python by applying z-score and IQR methods, illustrated with NumPy, Matplotlib, and Colab implementations.

Explore core probability concepts, including simple probability, sample space, and the addition rule with union and intersection, illustrated by coin toss, dice, and mutually exclusive events.

Explore conditional probability and its core formula P(A|B)=P(A∩B)/P(B). The lecture uses a dice example to show how many A outcomes lie in B.

Derive Bayes theorem from conditional probability by equating P(A and B) with P(B|A)P(A) and P(A|B)P(B), showing its standard and inverted forms.

Explore permutation and combination in probability and statistics for data science. Compare order matters in permutation with order doesn't matter in combination, using NPR and NCR formulas with factorials.

Explore the p-value, a probability between 0 and 1 that measures the likelihood of observed data under the null hypothesis, illustrated with spacebar and mouse pad examples.

Explore hypothesis testing by defining null and alternative hypotheses, performing four steps, and using a coin toss example to decide evidence via a significance level and rejection regions.

Explore type one and type two errors in hypothesis testing, and learn how the confusion matrix connects predictions to reality under null and alternative hypotheses.