5 animated scenes - sample space SIKUDNA, total probability tree, Bayes reverse, random variables, aur quiz.
"Diya gaya hai ki B ho chuka" = sample space ab sirf B! Die ke 6 outcomes dekho - B select karo (universe shrink hota hai), phir A select karo - B ke ANDAR A wale tiles glow. P(A|B) = P(A∩B)/P(B) LIVE.
Event A ke MULTIPLE raaste? P(A) = Σ P(Eᵢ)·P(A|Eᵢ) - har raaste ka (weight × uss raaste pe A ka chance) ADD karo. 2-bag tree LIVE: sliders se prior aur bag contents badlo, tree aur result update!
Defective item MIL GAYA - kis machine se aaya? P(Eᵢ|A) = P(Eᵢ)P(A|Eᵢ) ÷ Σ (denominator = total probability!). Sliders se machine output share aur defect rates badlo - PRIOR vs POSTERIOR bars compare karo.
X = outcome ka NUMBER (2 coins me heads: 0,1,2). Distribution table: ΣP = 1 zaroori! Mean E(X) = Σxᵢ·P(xᵢ) = long-run average. Coins ka count badlo - bars aur E(X) LIVE. (Variance/binomial syllabus se DELETED!)
5 sawaal. Yaad rakho: B diya to universe = B; independent ≠ mutually exclusive; total probability = raaston ka jod; Bayes ka denominator = total probability; E(X) = Σxᵢpᵢ!
P(A|B) = P(A∩B)/P(B), P(B)≠0 | B ke baad sample space = sirf B | P(A′|B) = 1 − P(A|B)
P(A∩B) = P(A)·P(B|A) | Independent: P(A∩B) = P(A)P(B) | Mutually exclusive ≠ INDEPENDENT (trap!) | Without replacement → draws NOT independent
E₁..Eₙ partition (disjoint, poora S) → P(A) = Σ P(Eᵢ)·P(A|Eᵢ) | Tree: branch pe multiply, across add!
P(Eᵢ|A) = P(Eᵢ)P(A|Eᵢ) / Σⱼ P(Eⱼ)P(A|Eⱼ) | Denominator = TOTAL PROBABILITY (half work!) | "A ho gaya - kis raaste se?" = posterior
X = outcome → number | Table: har pᵢ ≥ 0, Σpᵢ = 1 | E(X) = μ = Σxᵢ·pᵢ = long-run average | Variance/binomial DELETED
Bayes 5-marker = denominator pehle! | "Without replacement" → independence check | ΣP=1 verify karna aa sakta hai | Tree diagram = full marks ka rasta