Math ClubMath Club
v1 · padrão canônico

Lesson 76 — Normal distribution

Bell curve: density, Z-standardization, 68-95-99.7 rule, confidence intervals, and Z-tests. The central distribution of statistics and applied sciences.

Used in: Stochastik — Leistungskurs alemão · H2 Math — Singapura · AP Statistics — EUA · Math B — Japão

f(x)=1σ2πe(xμ)22σ2,Z=Xμσf(x) = \frac{1}{\sigma\sqrt{2\pi}}\,e^{-\frac{(x-\mu)^2}{2\sigma^2}}, \quad Z = \frac{X-\mu}{\sigma}
Choose your door

Rigorous notation, full derivation, hypotheses

Rigorous definition

Density and parameters

"If XX is a random variable and XX has a normal distribution with mean μ\mu and standard deviation σ\sigma, we write XN(μ,σ)X \sim N(\mu, \sigma). The mean μ\mu is the center of the symmetric curve, and the standard deviation σ\sigma gives the spread." — OpenStax Statistics §6.1

"Normal distributions are symmetric around their mean... The area under a normal distribution curve within one standard deviation of the mean is approximately 68%, within two standard deviations is approximately 95%, and within three standard deviations is approximately 99.7%." — OpenIntro Statistics §3.5

μμ−σμ+σμ−2σμ+2σ68%13.6%13.6%

Normal curve: 68% of data between μ ± σ (dark central region), 27.2% between μ ± 2σ (side regions), 0.3% in the tails beyond μ ± 3σ.

Solved examples

Exercise list

42 exercises · 10 with worked solution (25%)

Application 24Understanding 2Modeling 12Challenge 1Proof 3
  1. Ex. 76.1Application

    XN(70,102)X \sim \mathcal N(70, 10^2). Calculate the z-score for X=85X = 85.

  2. Ex. 76.2Application

    XN(100,152)X \sim \mathcal N(100, 15^2). Calculate the z-score for X=80X = 80.

  3. Ex. 76.3ApplicationAnswer key

    Calculate P(Z1,96)P(Z \leq 1{,}96).

  4. Ex. 76.4Application

    Calculate P(Z1,96)P(Z \geq 1{,}96).

  5. Ex. 76.5Application

    Calculate P(1,96Z1,96)P(-1{,}96 \leq Z \leq 1{,}96).

  6. Ex. 76.6Application

    Calculate P(Z1,5)P(Z \leq -1{,}5).

  7. Ex. 76.7Application

    Calculate P(0Z2)P(0 \leq Z \leq 2).

  8. Ex. 76.8Application

    XN(50,102)X \sim \mathcal N(50, 10^2). Calculate P(X>65)P(X > 65).

  9. Ex. 76.9Application

    XN(50,102)X \sim \mathcal N(50, 10^2). Calculate P(40<X<60)P(40 < X < 60).

  10. Ex. 76.10Application

    XN(0,4)X \sim \mathcal N(0, 4) (variance = 4). Calculate P(X>3)P(X > 3).

  11. Ex. 76.11ApplicationAnswer key

    Calculate the 90th percentile of N(100,152)\mathcal N(100, 15^2).

  12. Ex. 76.12Application

    Calculate Q1Q_1 (25th percentile) of N(100,152)\mathcal N(100, 15^2).

  13. Ex. 76.13Application

    IQ N(100,152)\sim \mathcal N(100, 15^2). What % of the population has an IQ between 85 and 115?

  14. Ex. 76.14Application

    IQ N(100,152)\sim \mathcal N(100, 15^2). What % has an IQ above 130?

  15. Ex. 76.15Application

    IQ N(100,152)\sim \mathcal N(100, 15^2). What % has an IQ above 145?

  16. Ex. 76.16Application

    Heights N(170,82)\sim \mathcal N(170, 8^2) cm. What % has a height above 186 cm?

  17. Ex. 76.17ApplicationAnswer key

    Grades N(70,102)\sim \mathcal N(70, 10^2). From which grade does the top 5% start?

  18. Ex. 76.18Application

    XN(0,1)X \sim \mathcal N(0, 1). Calculate P(X>3)P(|X| > 3).

  19. Ex. 76.19ApplicationAnswer key

    For XN(μ,σ2)X \sim \mathcal N(\mu, \sigma^2), what is the relationship between median, mode, and μ\mu?

  20. Ex. 76.20Application

    XN(20,42)X \sim \mathcal N(20, 4^2) and YN(10,32)Y \sim \mathcal N(10, 3^2) independent. What is the distribution of X+YX + Y?

  21. Ex. 76.21Application

    Monthly salary N(5000,15002)\sim \mathcal N(5000, 1500^2) reais. What is the salary floor for the top 10%?

  22. Ex. 76.22ApplicationAnswer key

    Flight duration N(120,102)\sim \mathcal N(120, 10^2) min. How much time to reserve to have 99% confidence of arriving on time?

  23. Ex. 76.23Application

    Daily stock returns N(0,001,  0,012)\sim \mathcal N(0{,}001,\; 0{,}01^2). Calculate P(loss>2%)P(\text{loss} > 2\%).

  24. Ex. 76.24Application

    Voltage N(220,52)\sim \mathcal N(220, 5^2). Device fails if V>235V > 235 V. Calculate the probability of failure.

  25. Ex. 76.25Modeling

    Parts with diameter N(10,00;  0,022)\mathcal N(10{,}00;\; 0{,}02^2) mm. Tolerance 10,00±0,0510{,}00 \pm 0{,}05 mm. What fraction is rejected?

  26. Ex. 76.26Modeling

    Survey with 1000 respondents estimates real proportion p=0,50p = 0{,}50. Construct a 95% CI for pp.

  27. Ex. 76.27ModelingAnswer key

    Xˉ=105\bar X = 105, n=25n = 25, σ=10\sigma = 10 (known). Construct a 95% CI for μ\mu.

  28. Ex. 76.28Modeling

    Test H0:μ=100H_0: \mu = 100 vs. H1:μ100H_1: \mu \neq 100. Xˉ=105\bar X = 105, n=25n = 25, σ=10\sigma = 10. Calculate the p-value and decide.

  29. Ex. 76.29ModelingAnswer key

    Execution time N(50,52)\sim \mathcal N(50, 5^2) ms. To guarantee SLA with 95% of requests below the limit, what threshold should be set?

  30. Ex. 76.30Modeling

    Six Sigma: specification μ±6σ\mu \pm 6\sigma. With a 1.5σ adjustment for process drift, calculate defects per million. Why is the result 3.4 ppm and not virtually zero?

  31. Ex. 76.31Modeling

    X-bar chart with n=5n = 5, σ=2\sigma = 2 (known), Xˉˉ=100\bar{\bar{X}} = 100. Calculate UCL and LCL at ±3σ\pm 3\sigma.

  32. Ex. 76.32Modeling

    ML model scores N(0,80,  0,052)\sim \mathcal N(0{,}80,\; 0{,}05^2). What is the threshold to select the top 20% of models?

  33. Ex. 76.33Modeling

    100Ω resistor with ± 5% tolerance. Assuming σ=5/3\sigma = 5/3 ohm (3σ\sigma = tolerance), calculate the fraction within specification.

  34. Ex. 76.34Modeling

    Annual portfolio return N(5%,20%2)\sim \mathcal N(5\%, 20\%^2). Calculate the probability of a negative return in one year.

  35. Ex. 76.35Modeling

    ENEM grades (Math) N(520,1102)\sim \mathcal N(520, 110^2). Calculate P(grade>700)P(\text{grade} > 700).

  36. Ex. 76.36ModelingAnswer key

    Annual IPCA modeled as N(4,5%,  1,5%2)\mathcal N(4{,}5\%,\; 1{,}5\%^2). Inflation target: up to 6.5%. What is the probability of exceeding the target?

  37. Ex. 76.37Understanding

    Why do we standardize to the standard normal? What justifies the existence of a single Φ(z)\Phi(z) table?

  38. Ex. 76.38Understanding

    Is the normal tail "thin" or "heavy"? Why does this matter in financial risk modeling?

  39. Ex. 76.39Challenge

    Show that if XN(0,1)X \sim \mathcal N(0, 1), then Y=X2Y = X^2 has a chi-squared distribution with 1 degree of freedom.

  40. Ex. 76.40Proof

    Demonstrate that if XN(μ,σ2)X \sim \mathcal N(\mu, \sigma^2) and Y=aX+bY = aX + b (with a>0a > 0), then YN(aμ+b,  a2σ2)Y \sim \mathcal N(a\mu + b,\; a^2\sigma^2).

  41. Ex. 76.41ProofAnswer key

    Demonstrate that +ex2/2dx=2π\int_{-\infty}^{+\infty} e^{-x^2/2}\,dx = \sqrt{2\pi} using the polar coordinates trick.

  42. Ex. 76.42ProofAnswer key

    Demonstrate (sketch) that the normal distribution maximizes differential entropy among all continuous distributions with fixed mean μ\mu and variance σ2\sigma^2.

Sources

  • OpenIntro Statistics (4th ed) — Diez, Çetinkaya-Rundel, Barr · 2019 · EN · CC-BY-SA. Primary source — §3.5 (standardization, 68-95-99.7 rule, Q-Q plot, applications).
  • Statistics (OpenStax) — Illowsky, Dean · 2022 · EN · CC-BY. §6.1–6.4 — density, CDF, CI, CLT, AP-level exercises.
  • Introduction to Probability (Grinstead-Snell) — Grinstead, Snell · 1997 · EN · GNU FDL. §5.2 — Gaussian integral, MGF, maximum entropy, De Moivre-Laplace limit; demonstrative exercises.

Updated on 2025-05-14 · Author(s): Clube da Matemática

Found an error? Open an issue on GitHub or submit a PR — open source forever.