4.9.11
Intuition Bada picture (KYUN)Do random variables independent hote hain jab ek ki value jaanne se doosre ke baare mein kuch bhi pata nahi chalta. Poora point yeh hai ki joint behaviour, separate behaviours mein factorise ho jaata hai — tum har variable ko akele study karke bas multiply kar sakte ho. Independence woh assumption hai jo mushkil joint problems ko aasaan product problems mein badal deti hai.
Intuition Events se kyun shuru karein?Tum pehle se jaante ho ki do events A , B P ( A ∩ B ) = P ( A ) P ( B ) X Y har baar independent ho.
Definition Random variables ki independence (general)Random variables X Y independent hain agar saare (Borel) sets A , B ⊆ R P ( X ∈ A , Y ∈ B ) = P ( X ∈ A ) P ( Y ∈ B ) . { X ∈ A } { Y ∈ B } A B
YEH KYA KEHTA HAI: plane ke kisi bhi joint region ki probability, marginal probabilities ke product mein split ho jaati hai.
Special sets A = ( − ∞ , x ] B = ( − ∞ , y ] P ( X ≤ x , Y ≤ y ) = P ( X ≤ x ) P ( Y ≤ y ) .
Left side wahi hai joint CDF F X , Y ( x , y ) F X ( x ) F Y ( y )
Yeh kaafi kyun hai (sirf necessary nahi)? ( − ∞ , x ] generate karte hain, toh agar product rule inpe hold karta hai toh yeh har A , B
Intuition Teeno ke liye ek slogan:joint object marginals ke product mein factor hota hai. CDF, PMF, PDF — same idea, alag language.
Worked example Independence ek nazar mein pehchannaf X , Y ( x , y ) = 4 x y [ 0 , 1 ] 2
Kyun split karein: 4 x y = ( 2 x ) ( 2 y ) Toh X ⊥ Y f X ( x ) = 2 x , f Y ( y ) = 2 y
Worked example Discrete checkJoint PMF table:
X \ Y 0
1
0
0.12
0.28
1
0.18
0.42
Step 1 — marginals. Kyun? Independence marginals ke relative define hoti hai.
p X ( 0 ) = 0.12 + 0.28 = 0.40 p X ( 1 ) = 0.60 p Y ( 0 ) = 0.30 p Y ( 1 ) = 0.70 Step 2 — ek cell test karo. p X ( 0 ) p Y ( 0 ) = 0.40 ⋅ 0.30 = 0.12 = p X , Y ( 0 , 0 ) Step 3 — saari cells test karo. 0.40 ⋅ 0.70 = 0.28 0.60 ⋅ 0.30 = 0.18 0.60 ⋅ 0.70 = 0.42 Saari cells kyun? Ek akela passing cell kaafi nahi; independence har jagah hold karni chahiye.
→ X ⊥ Y
Worked example Continuous failure (support coupling)f X , Y ( x , y ) = 2 0 ≤ x ≤ y ≤ 1
Kyun suspicious hai: support khud x y x ≤ y X = 0.9 Y ≥ 0.9 Marginals: f X ( x ) = ∫ x 1 2 d y = 2 ( 1 − x ) f Y ( y ) = ∫ 0 y 2 d x = 2 y
Product f X f Y = 4 y ( 1 − x ) = 2 NOT independent , chahe "2" separable lagta ho. ✅ lesson learned.
Worked example Independence via factorisationf X , Y ( x , y ) = e − x − y x , y > 0
Kyun independent: = e − x ⋅ e − y ( 0 , ∞ ) 2 X , Y ∼ Exp ( 1 )
Common mistake Steel-man: "Zero covariance ⇒ independent"Kyun sahi lagta hai: independence Cov = 0 Kyun galat hai: Cov = 0 linear dependence ko khatam karta hai. Maano X ∼ U ( − 1 , 1 ) Y = X 2 E [ X Y ] = E [ X 3 ] = 0 = E X E Y Y completely determined hai X Fix: independence ⇒ kabhi nahi ulta (except jointly Gaussian, jahan dono coincide karte hain).
Common mistake Steel-man: "Separable formula hamesha ⇒ independent"Kyun sahi lagta hai: PDF g ( x ) h ( y ) jaisa dikhta hai.
Kyun galat hai: support bhi rectangle (product set) hona chahiye. Upar wala triangle example bilkul yahi pe fail karta hai.
Fix: check karo ki dono density factor bhi kare aur support A × B
Recall Feynman: 12-saal ke bachche ko explain karoSocho fair mein do spinning wheels hain. Agar woh independent hain, toh pehli wheel ka red pe aana doosri wheel ke baare mein bilkul kuch nahi bataata. "Pehli red AND doosri blue" ki probability nikaalene ke liye tum bas dono ki probability alag-alag multiply kar dete ho. Yahi multiply-karne-waali trick saari baat hai — aur yeh har pair of outcomes ke liye kaam karta hai, sirf ek lucky pair ke liye nahi.
"JOINT = PRODUCT, EVERYWHERE."
Agar joint har point pe marginals mein factor ho jaaye (aur support ek box ho), toh woh independent hain. Warna, woh gossip karte hain.
Events ke terms mein X , Y Sabhi Borel A , B P ( X ∈ A , Y ∈ B ) = P ( X ∈ A ) P ( Y ∈ B )
Independence ke liye CDF criterion. F X , Y ( x , y ) = F X ( x ) F Y ( y ) x , y
Continuous-case criterion. f X , Y ( x , y ) = f X ( x ) f Y ( y )
Discrete-case criterion. p X , Y ( x , y ) = p X ( x ) p Y ( y ) x , y
CDF test sirf necessary nahi balki sufficient kyun hai? Intervals ( − ∞ , x ] A , B
Factorisation theorem condition. Joint = g ( x ) h ( y )
Kya f = 2 0 ≤ x ≤ y ≤ 1 Nahi — support x , y = 2
Independence expectation ke baare mein kya kehti hai? E [ X Y ] = E [ X ] E [ Y ] Cov = 0
Kya Cov = 0 Nahi (e.g. Y = X 2 X ∼ U ( − 1 , 1 )
Formula se independence confirm karne ke liye do extra conditions. Density factor kare AUR support ek product (rectangle) set ho.
Joint distribution and marginals Conditional distributions and conditional independence Covariance and correlation Jointly Gaussian random variables Sums of independent random variables — convolution Independence of events
holds for all Borel sets A,B
special sets minus inf to x
intervals generate all Borel sets
normalising constant cancels
Event independence P of A and B = P A P B
RV independence general def
Joint prob factors into marginals
Joint object = product of marginals
Condition is sufficient not just necessary
Factorisation theorem g x times h y
Spot independence instantly