DISTRIBUCIONES de PROBABILIDAD: Todo lo que TIENES que SABER

Main ideas / concepts covered

1. Probability distributions and random variables

   • The video introduces random variables as quantities defined from random experiments.
   • Examples:
     • Discrete random variable: the number of sixes obtained when rolling 10 dice (values (0) to (10)).
     • Continuous random variable: a student’s height (values over an interval, not just countable outcomes).
   • Two main categories:
     • Discrete variables: take a countable set of distinct values.
     • Continuous variables: take values over an interval (uncountably many), so the probability of being exactly equal to a single point is 0.

2. Binomial distribution (for discrete variables)

   • The binomial distribution models discrete outcomes like:
     • the number of successes in (n) independent trials
     • where each trial has the same success probability (p)
   • The video focuses on the dice example (success = getting a six).
   • Binomial parameters:
     • (n) = number of trials (e.g., 10 dice)
     • (p) = probability of success on one trial (e.g., (p = 1/6))

3. Computing binomial probabilities

   • Uses the binomial probability formula to compute (P(X=k)).
   • For probabilities involving inequalities (e.g., (X>3)):
     • Rewrite them as unions of events (X=4,5,\dots,n).
   • Two equivalent approaches:
     • Direct summation of relevant probabilities
     • Complement rule: (P(\text{event}) = 1 - P(\text{complement}))

4. Connection to the binomial theorem

   • The binomial probability expression is shown to resemble the binomial theorem expansion of ((a+b)^n).
   • The total probability over (k=0) to (n) equals 1, matching the full sample space.

5. Expected value, variance, and standard deviation (binomial case)

   • Expected value (mean):

     • [ E[X]=np ]

   • Variance:

     • [ \mathrm{Var}(X)=np(1-p) ]

   • Standard deviation:

     • [ \sigma=\sqrt{np(1-p)} ]

   • Interpretation: variance/standard deviation quantify dispersion around the mean.

6. Bernoulli distribution (special case of binomial)

   • A Bernoulli distribution occurs when (n=1).
   • With parameter (p), the outcomes are:
     • (0) with probability (1-p)
     • (1) with probability (p)

7. Normal distribution (for continuous variables)

   • Introduces the normal distribution as a common continuous model.
   • Parameters:
     • (\mu) = mean (expected value)
     • (\sigma) = standard deviation (dispersion)
   • Probability geometry:
     • The normal PDF does not give (P(X=K)) at a single point.
     • Probabilities are areas under the curve:
       • (P(X \le K)) corresponds to area from (-\infty) to (K).
   • Key fact: for continuous variables, (P(X=K)=0) (zero-width area).

8. Standardization (using Z and the standard normal table)

   • Define:

     • [ Z=\frac{X-\mu}{\sigma} ]

   • Then (Z) follows the standard normal distribution (mean (0), standard deviation (1)).

   • Tables provide values such as (P(Z\le z)).

9. Properties and symmetry of the normal distribution

   • Uses symmetry to convert left-tail probabilities to right-tail equivalents.
   • Uses the complement rule:
     • (P(X>a)=1-P(X\le a))
   • For intervals:
     • [ P(a\le X\le b)=P(X\le b)-P(X\le a) ]

10. Approximation: binomial by normal (with continuity correction)

    • Explains when/how to approximate a binomial distribution with a normal one.
    • Example: manufacturing 1000 pieces with defect probability (0.05)
      • Let (X) = number of defective pieces:
        • (X \sim \mathrm{Binomial}(n=1000,\;p=0.05))
    • Motivation:
      • Exact binomial probabilities like (P(X<50)) require summing many terms.
    • Approximation:
      • Approximate (X) by a normal variable (Y):
        • mean (\mu=np)
        • standard deviation (\sigma=\sqrt{np(1-p)})
    • Continuity correction:
      • If approximating (P(X<50)), use (P(Y\le 49.5)).
    • Then compute using standardization + a Z-table.

---

Methodologies / instruction-like steps (detailed)

A) Classifying the random variable

• Given a random experiment, define a variable (X).
• Determine the type:
  • Discrete: values can be listed as a countable set (e.g., number of sixes (0) to (10)).
  • Continuous: values fall in an interval (e.g., heights).
• Choose distribution family accordingly:
  • Binomial → discrete
  • Normal → continuous

B) Binomial probability (P(X=k))

• Identify:
  • (n) = number of trials
  • (p) = probability of success per trial

• Use:

  • [ P(X=k)=\binom{n}{k}p^{k}(1-p)^{n-k} ]

• Where:

  • [ \binom{n}{k}=\frac{n!}{k!(n-k)!} ]

• Substitute (n, p, k) and compute.

C) Binomial probability for inequalities (e.g., (P(X>3)))

• Rewrite the inequality as a union of point events:
  • If (P(X>3)), then (X \in {4,5,\dots,n}).

• Compute using either:

  • Direct summation:

    • [ P(X>3)=\sum_{i=4}^{n}P(X=i) ]

  • Complement rule:

    • [ P(X>3)=1-P(X\le 3) ]

    • with:

      • [ P(X\le 3)=\sum_{i=0}^{3}P(X=i) ]

D) Expected value and variance for binomial

• For (X \sim \mathrm{Binomial}(n,p)):

  • [ E[X]=np ]

  • [ \mathrm{Var}(X)=np(1-p) ]

  • [ \sigma=\sqrt{np(1-p)} ]

E) Normal distribution probabilities via standardization

• For (X \sim \mathcal{N}(\mu,\sigma^2)) and threshold (K):

  1. Convert to standard normal:

     • [ Z=\frac{X-\mu}{\sigma} ]

  2. Transform:

     • [ P(X\le K)=P\left(Z\le \frac{K-\mu}{\sigma}\right) ]

  3. Look up the value in the standard normal (Z) table. - Use rules: - Complement:

     • (P(X>K)=1-P(X\le K))
     • Between two values:
     • [ P(a\le X\le b)=P(X\le b)-P(X\le a) ]

• Handle left-tail cases using symmetry.

F) Approximating binomial with normal + continuity correction

• If (X \sim \mathrm{Binomial}(n,p)), approximate with (Y\sim \mathcal{N}(\mu,\sigma^2)) where:

  • [ \mu=np ]

  • [ \sigma=\sqrt{np(1-p)} ]

• Apply continuity correction:

  • For example, approximate (P(X<50)) by (P(Y\le 49.5)).
• Then compute:
  • Convert the cutoff (c) to a Z-score (\frac{c-\mu}{\sigma})
  • Use the Z-table for (P(Y\le c)).

---

Speakers / sources featured

• No other named speakers or external sources are shown in the subtitles.
• Mentions:
  • Newton’s binomial theorem (referenced as a concept/source attributed to Newton, not as a spoken guest).