The Normal Distribution -...
Transcript of The Normal Distribution -...
![Page 1: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/1.jpg)
The Normal DistributionThe Normal Distribution
Will MonroeJuly 19, 2017
with materials byMehran Sahamiand Chris Piech
image: Etsy
![Page 2: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/2.jpg)
Announcements: Midterm
A week from yesterday:
Tuesday, July 25, 7:00-9:00pmBuilding 320-105
One page (both sides) of notes
Material through today’s lecture
Review session:Tomorrow, July 20, 2:30-3:20pmin Gates B01
![Page 3: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/3.jpg)
Review: A grid of random variables
X∼Geo(p)
number of successes time to get successes
Onetrial
Severaltrials
Intervalof time X∼Exp(λ)
Onesuccess
Severalsuccesses
One success after interval
of time
X∼NegBin (r , p)
X∼Ber(p)
X∼Bin(n , p)
X∼Poi(λ)
n = 1
Onesuccess
Onesuccess
r = 1
(continuous!)
![Page 4: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/4.jpg)
Review: Continuous distributions
A continuous random variable has a value that’s a real number (not necessarily an integer).
Replace sums with integrals!
P(a<X≤b)=F X (b)−F X (a)
F X (a)= ∫x=−∞
a
dx f X (x)
![Page 5: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/5.jpg)
Review: Probability density function
The probability density function (PDF) of a continuous random variable represents the relative likelihood of various values.
Units of probability divided by units of X. Integrate it to get probabilities!
P(a<X≤b)=∫x=a
b
dx f X (x)
![Page 6: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/6.jpg)
Continuous expectation and variance
Remember: replace sums with integrals!
E [X ]= ∫x=−∞
∞
dx x⋅f X (x)E [X ]= ∑x=−∞
∞
x⋅pX (x)
E [X2]= ∑
x=−∞
∞
x2⋅pX (x) E [X2
]= ∫x=−∞
∞
dx x2⋅f X (x)
Var(X )=E [(X−E [X ])2]=E [X2
]−(E [X ])2
(still!)
![Page 7: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/7.jpg)
Review: Uniform random variable
A uniform random variable is equally likely to be any value in a single real number interval.
f X (x)={1
β−αif x∈[α ,β]
0 otherwise
X∼Uni(α ,β)
![Page 8: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/8.jpg)
Uniform: Fact sheet
PDF:
expectation: E [X ]=α+β
2
variance: Var(X )=(β−α)
2
12
maximum value
minimum value
X∼Uni(α ,β)
f X (x)={1
β−αif x∈[α ,β]
0 otherwise
CDF: F X (x)={x−αβ−α
if x∈[α ,β]
1 if x>β
0 otherwise
image: Haha169
![Page 9: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/9.jpg)
Review: Exponential random variable
An exponential random variable is the amount of time until the first event when events occur as in the Poisson distribution.
f X (x)={λ e−λ x if x≥00 otherwise
X∼Exp(λ)
image: Adrian Sampson
![Page 10: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/10.jpg)
Exponential: Fact sheet
PDF:
expectation: E [X ]=1λ
variance: Var(X )=1λ
2
time until first event
rate of events per unit time
X∼Exp(λ)
CDF:
image: Adrian Sampson
f X (x)={λ e−λ x if x≥00 otherwise
F X (x)={1−e−λ x if x≥00 otherwise
![Page 11: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/11.jpg)
Normal random variable
An normal (= Gaussian) random variable is a good approximation to many other distributions. It often results from sums or averages of independent random variables.
X∼N (μ ,σ2)
f X (x)=1
σ √2πe
−12 ( x−μ
σ )2
![Page 12: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/12.jpg)
Déjà vu?
![Page 13: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/13.jpg)
Déjà vu?
P(X=k )
k
![Page 14: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/14.jpg)
Déjà vu?
f X (x)
x
X = sum of n independent Uni(0, 1) variables
image: Thomasda
![Page 15: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/15.jpg)
“The normal distribution”
Also known as: Gaussian distribution
Shape: bell curve
Personality: easygoing
![Page 16: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/16.jpg)
What is normally distributed?
Natural phenomena: heights, weights…
Noise in measurements
Sums/averages of many random variables
Averages of samples from a population
(approximately)
(caveats: independence,equal weighting,continuity...)
(with sufficient sample sizes)
![Page 17: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/17.jpg)
The Know-Nothing Distribution“maximum entropy”
The normal is the most spread-out distribution with a fixed expectation and variance.
If you know E[X] and Var(X) but nothing else,a normal is probably a good starting point!
![Page 18: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/18.jpg)
Normal: Fact sheet
PDF:
mean
X∼N (μ ,σ2)
f X (x)=1
σ √2πe
−12 ( x−μ
σ )2
variance (σ = standard deviation)
![Page 19: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/19.jpg)
The Standard Normal
Z∼N (0,1)
μ σ²
X∼N (μ ,σ2) X=σ Z+μ
Z=X−μσ
![Page 20: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/20.jpg)
De-scarifying the normal PDF
f X (x)=1
σ √2πe
−12 ( x−μ
σ )2
![Page 21: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/21.jpg)
De-scarifying the normal PDF
f Z (z)=1
1√2πe
−12 ( z−0
1 )2
![Page 22: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/22.jpg)
De-scarifying the normal PDF
f Z (z)=1
√2πe
−12
z2
![Page 23: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/23.jpg)
De-scarifying the normal PDF
f Z (z)=C e−
12
z2
![Page 24: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/24.jpg)
De-scarifying the normal PDF
f Z (z)=C e−
12
z2
−12
z2
![Page 25: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/25.jpg)
De-scarifying the normal PDF
f Z (z)=C e−
12
z2
−12
z2
![Page 26: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/26.jpg)
De-scarifying the normal PDF
f X (x)=1
σ √2πe
−12 ( x−μ
σ )2
normalizingconstant
Z=X−μσ
![Page 27: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/27.jpg)
Normal: Fact sheet
PDF:
mean
X∼N (μ ,σ2)
CDF:
f X (x)=1
σ √2πe
−12 ( x−μ
σ )2
F X (x)=Φ ( x−μσ )=∫
−∞
x
dx f X (x)
variance (σ = standard deviation)
(no closed form)
![Page 28: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/28.jpg)
The Standard Normal
Z∼N (0,1)
μ σ²
X∼N (μ ,σ2) X=σ Z+μ
Z=X−μσ
Φ(z)=FZ(z)=P(Z≤z)
![Page 29: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/29.jpg)
Symmetry of the normal
P(X≤μ−x)=P(X≥μ+x)
and don’t forget:
P(X>x)=1−P(X≤x)
![Page 30: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/30.jpg)
Symmetry of the normal
P(Z≤−z)=P(Z≥z)
and don’t forget:
P(Z> z)=1−P(Z≤z)
![Page 31: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/31.jpg)
Symmetry of the normal
Φ(−z)=P(Z≥z)
and don’t forget:
P(Z> z)=1−Φ(z)
![Page 32: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/32.jpg)
The standard normal table
Φ(0.54)=P(Z≤0.54)=0.7054
![Page 33: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/33.jpg)
With today’s technology
scipy.stats.norm(mean, std).cdf(x)
standard deviation! not variance.you might need math.sqrt here.
![Page 34: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/34.jpg)
Break time!
![Page 35: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/35.jpg)
Practice with the Gaussian
X ~ N(3, 16) μ = 3 σ² = 16 σ = 4
P(X>0)=P( X−34
>0−3
4 )=P (Z>−
34 )
=1−P(Z≤−34 )=1−Φ(−
34)
=1−(1−Φ(34))
=Φ(34)≈0.7734
![Page 36: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/36.jpg)
Practice with the Gaussian
X ~ N(3, 16) μ = 3 σ² = 16 σ = 4
P(|X−3|>4)=P (X <−1)+P(X>7)
=P ( X−34
<−1−3
4 )+P ( X−34
>7−3
4 )=P (Z<−1)+P(Z>1)
=Φ(−1)+(1−Φ(1))
=(1−Φ(1))+(1−Φ(1))
≈2⋅(1−0.8413)
=0.3173
![Page 37: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/37.jpg)
Practice with the Gaussian
X ~ N(3, 16) μ = 3 σ² = 16 σ = 4
P(|X−μ|>σ)=P(X<μ−σ)+P(X>μ+σ)
=P ( X−μσ <
μ−σ−μσ )+P ( X−μ
σ >μ+σ−μ
σ )=P (Z<−1)+P(Z>1)
=Φ(−1)+(1−Φ(1))
=(1−Φ(1))+(1−Φ(1))
≈2⋅(1−0.8413)
=0.3173
![Page 38: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/38.jpg)
Normal: Fact sheet
PDF:
expectation: E [X ]=μ
variance: Var(X )=σ2
mean
X∼N (μ ,σ2)
CDF:
f X (x)=1
σ √2πe
−12 ( x−μ
σ )2
F X (x)=Φ ( x−μσ )=∫
−∞
x
dx f X (x)
variance (σ = standard deviation)
(no closed form)
![Page 39: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/39.jpg)
Carl Friedrich Gauss
(1775-1855)—remarkably influential German mathematician
Started doing groundbreaking math as a teenager
Didn’t invent the normal distribution (but popularized it)
![Page 40: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/40.jpg)
Noisy wires
Send a voltage of X = 2 or -2 on a wire.+2 represents 1, -2 represents 0.
Receive voltage of X + Y on other end,where Y ~ N(0, 1).
If X + Y ≥ 0.5, then output 1, else 0.
P(incorrect output | original bit = 1) =
P(2+Y <0.5)=P (Y <−1.5)
=Φ(−1.5)
=1−Φ(1.5)≈0.0668
![Page 41: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/41.jpg)
Noisy wires
Send a voltage of X = 2 or -2 on a wire.+2 represents 1, -2 represents 0.
Receive voltage of X + Y on other end,where Y ~ N(0, 1).
If X + Y ≥ 0.5, then output 1, else 0.
P(incorrect output | original bit = 0) =
P(−2+Y≥0.5)=P(Y≥2.5)
=1−P(Y <2.5)
=1−Φ(2.5)≈0.0062
![Page 42: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/42.jpg)
Poisson approximation to binomial
P(X=k )
k
large n, small p
Bin (n , p)≈Poi (λ)
![Page 43: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/43.jpg)
Normal approximation to binomial
P(X=k )
k
large n, medium p
Bin (n , p)≈N (μ ,σ2)
![Page 44: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/44.jpg)
Something is strange...
![Page 45: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/45.jpg)
Continuity correction
When approximating a discrete distribution with a continuous distribution, adjust the bounds by 0.5 to account for the missing half-bar.
P(X≥55)≈P(Y >54.5)
X∼Bin (n , p)
Y∼N (np ,np(1−p))
![Page 46: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/46.jpg)
Miracle diets100 people placed on a special diet.
Doctor will endorse diet if ≥ 65 people have cholesterol levels decrease.
What is P(doctor endorses | diet has no effect)?
X: # people whose cholesterol decreases
X ~ Bin(100, 0.5) np = 50 np(1 – p) = 50(1 – 0.5) = 25
≈ Y ~ N(50, 25)
P(Y >64.5)=P(Y−505
>64.5−50
5 )=P(Z>2.9)=1−Φ(2.9)≈0.00187
![Page 47: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/47.jpg)
Stanford admissions
P(Y >1749.5)=P(Y−1686.4
√539.65>
1749.5−1686.4
√539.65 )≈P(Z>2.54)=1−Φ(2.54)≈0.0053
image: Victor Gane
Stanford accepts 2480 students.Each student independently decides to attend with p = 0.68.
What isP(at least 1750 students attend)?
X: # of students who will attend.X ~ Bin(2480, 0.68) np = 1686.4 σ² = np(1 – p) ≈ 539.65
≈ Y ~ N(1686.4, 539.65)
![Page 48: The Normal Distribution - web.stanford.eduweb.stanford.edu/class/archive/cs/cs109/cs109.1178/lectures/11-normal-distribution.pdf · The Normal Distribution Will Monroe July 19, 2017](https://reader035.fdocuments.us/reader035/viewer/2022070617/5d547d1988c9932d4a8bcee7/html5/thumbnails/48.jpg)
Stanford admissions changes