Probability Review for Final Exam

18.05 Spring 2014

Jeremy Orloff and Jonathan Bloom

Problem 1. We compute
[
1 2 3]15 + 1 · 4 5 2
E[X] = −2 · + −1 · +0· +2· = .
15 15 15 15 3
Thus
2 14
Var(X) = E((X − )2 ) = .
3 9

Problem 2. We first compute

1
2
E[X] = x · 2xdx =
0 3
1
1
E[X 2 ] = x2 · 2xdx =
0 2
1
1
E[X 4 ] = x4 · 2xdx = .
0 3
Thus,
1 4 1
Var(X) = E[X 2 ] − (E[X])2 = − =
2 9 18
and
2 1 1 1
Var(X 2 ) = E[X 4 ] = E[X 2 ] = − = .
3 4 12

Problem 3. Use Var(X) = E(X 2 ) − E(X)2 ⇒ 3 = E(X 2 ) − 4 ⇒ E(X 2 ) = 7.

Problem 4. answer:
Make a table X: 0 1
prob: (1-p) p
X2 0 1.
From the table, E(X) = 0 · (1 − p) + 1 · p = p.
Since X and X 2 have the same table E(X 2 ) = E(X) = p.
Therefore, Var(X) = p − p2 = p(1 − p).

Problem 5. Let X be the number of people who get their own hat.
Following the hint: let Xj represent whether person j gets their own hat. That is,

1
Xj = 1 if person j gets their hat and 0 if not.
100
1 100
1
We have, X = Xj , so E(X) = E(Xj ).
j=1 j=1
Since person j is equally likely to get any hat, we have P (Xj = 1) = 1/100. Thus,
Xj ∼ Bernoulli(1/100) ⇒ E(Xj ) = 1/100 ⇒ E(X) = 1.

Problem 6. (a) There are a number of ways to present this.

X ∼ 3 binomial(25, 1/6), so
k 25−k
25 1 5
P (X = 3k) = , for k = 0, 1, 2, . . . , 25.
k 6 6

(b) X ∼ 3 binomial(25, 1/6).

Recall that the mean and variance of binomial(n, p) are np and np(1 − p). So,

E(X) = 3 E(textbinomial(25, 1/6)) = 3·25/6 = 75/6, and Var(X) = 9 Var(textbinomial(25, 1/6)) = 9·25(

(c) E(X + Y ) = E(X) + E(Y ) = 150/6 = 25., E(2X) = 2E(X) = 150/6 = 25.
Var(X + Y ) = Var(X) + Var(Y ) = 250/4. Var(2X) = 4Var(X) = 500/4.
The means of X + Y and 2X are the same, but Var(2X) > Var(X + Y ).
This makes sense because in X +Y sometimes X and Y will be on opposite sides from
the mean so distances to the mean will tend to cancel, However in 2X the distance
to the mean is always doubled.

Problem 7. First we find the value of a:

Z 1 Z 1
1 a
f (x) dx = 1 = x + ax2 dx = + ⇒ a = 3/2.
0 0 2 3

The CDF is FX (x) = P (X ≤ x). We break this into cases:

(i) b < 0 ⇒ FX (b) = 0.
Z b
3 b2 b3
(ii) 0 ≤ b ≤ 1 ⇒ FX (b) = x + x2 dx = + .
0 2 2 2
(iii) 1 < x ⇒ FX (b) = 1.
Using FX we get

.52 + .53 13
P (.5 < X < 1) = FX (1) − FX (.5) = 1 − = .
2 16

Problem 8. (i) yes, discrete, (ii) no, (iii) no, (iv) no, (v) yes, continuous

2
(vi) no (vii) yes, continuous, (viii) yes, continuous.

Problem 9. (a) We compute

Z 5
P (X ≥ 5) = 1 − P (X < 5) = 1 − λe−λx dx = 1 − (1 − e−5λ ) = e−5λ .
0

(b) We want P (X ≥ 15|X ≥ 10). First observe that P (X ≥ 15, X ≥ 10) = P (X ≥

15). From similar computations in (a), we know

P (X ≥ 15) = e−15λ P (X ≥ 10) = e−10λ .

From the definition of conditional probability,

P (X ≥ 15, X ≥ 10) P (X ≥ 15)

P (X ≥ 15|X ≥ 10) = = = e−5λ
P (X ≥ 10) P (X ≥ 10)

Note: This is an illustration of the memorylessness property of the exponential

distribution.

Problem 10.
(a) We did this in class. Let φ(z) and Φ(z) be the PDF and CDF of Z.
FY (y) = P (Y ≤ y) = P (aZ + b ≤ y) = P (Z ≤ (y − b)/a) = Φ((y − b)/a).
Differentiating:
d d 1 1 2 2
fY (y) = FY (y) = Φ((y − b)/a) = φ((y − b)/a) = √ e−(y−b) /2a .
dy dy a 2π a

Since this is the density for N(b, a2 ) we have shown Y ∼ N(b, a2 ).

(b) By part (a), Y ∼ N(µ, σ 2 ) ⇒ Y = σZ + µ.
But, this implies (Y − µ)/σ = Z ∼ N(0, 1). QED

Problem 11. (a) E(W ) = 3E(X) − 2E(Y ) + 1 = 6 − 10 + 1 = −3

Var(W ) = 9Var(X) + 4Var(Y ) = 45 + 36 = 81
(b) Since the sum of independent normal is normal part  W ∼ N (−3, 81).
 (a) shows:
W +3 9
Let Z ∼ N (0, 1). We standardize W : P (W ≤ 6) = P ≤ = P (Z ≤ 1) ≈ .84.
9 9

Problem 12.
Method 1
1
U (a, b) has density f (x) = on [a, b]. So,
b−a

3
 b b
�b
x2 �� b2 − a 2
Z Z
1 a+b
E(X) = xf (x) dx = x dx = = = .
a a b−a 2(b − a) a 2(b − a)
� 2
Z b Z b �b
2 2 1 2 x3 �� b3 − a3
E(X ) = x f (x) dx = x dx = = .
a b−a a 3(b − a) �a 3(b − a)

Finding Var(X) now requires a little algebra,

b3 − a3 (b + a)2
Var(X) = E(X 2 ) − E(X)2 = −
3(b − a) 4
4(b3 − a3 ) − 3(b − a)(b + a)2 b3 − 3ab2 + 3a2 b − a3 (b − a)3 (b − a)2
= = = = .
12(b − a) 12(b − a) 12(b − a) 12

Method 2
There is an easier way to find E(X) and Var(X).
Let U ∼ U(a, b). Then the calculations above show E(U ) = 1/2 and (E(U 2 ) = 1/3
⇒ Var(U ) = 1/3 − 1/4 = 1/12.
Now, we know X = (b−a)U +a, so E(X) = (b−a)E(U )+a = (b−a)/2+a = (b+a)/2
and Var(X) = (b − a)2 Var(U ) = (b − a)2 /12.

Problem 13.
(a) Sn ∼ Binomial(n, p), since it is the number of successes in n independent
Bernoulli trials.
(b) Tm ∼ Binomial(m, p), since it is the number of successes in m independent
Bernoulli trials.
(c) Sn + Tm ∼ Binomial(n + m, p), since it is the number of successes in n + m
independent Bernoulli trials.
(d) Yes, Sn and Tm are independent. We haven’t given a formal definition of
independent random variables yet. But, we know it means that knowing Sn gives no
information about Tm . This is clear since the first n trials are independent of the last
m.

Problem 14. Compute the median for the exponential distribution with parameter
λ. The density for this distribution is f (x) = λ e−λx . We know (or can compute)
that the distribution function is F (a) = 1 − e−λa . The median is the value of a such
that F (a) = .5. Thus, 1 − e−λa = 0.5 ⇒ 0.5 = e−λa ⇒ log(0.5) = −λa ⇒
a = log(2)/λ.

4
Problem 15. (a) The joint distribution is given by

Y \X 1 2 3
1168 825 305 2298
1 5383 5383 5383 5383

573 1312 1200 3085

2 5383 5383 5383 5383
1741 2137 1505
5383 5383 5383
1

with the marginal distribution of X at right and of Y at bottom.

(b) X and Y are dependent because, for example,
1168
P (X = 1 and Y = 1) =
5383
is not equal to
1741 2298
P (X = 1)P (Y = 1) = · .
5383 5383

Problem 16. (a) Here we have two continuous random variables X and Y with
going potability density function
12
f (x, y) = xy(1 + y) for 0 ≤ x ≤ 1 and 0 ≤ y ≤ 1,
5
and f (x, y) = 0 otherwise. So
Z 1 Z 2
1 1 1 2 2 3 41
P( ≤ X ≤ , ≤ Y ≤ ) = f (x, y)dy dx = .
4 2 3 3 1
4
1
3
720

a b
(b) F (a, b) = 0 0
f (x, y)dy dx = 53 a2 b2 + 52 a2 b3 for 0 ≤ a ≤ 1 and 0 ≤ b ≤ 1.
(c) Since f (x, y) = 0 for y > 1, we have

FX (a) = lim F (a, b) = F (a, 1) = a2 .

b→∞

(d) For 0 ≤ x ≤ 1, we have

Z ∞ Z 1
fX (x) = f (x, y)dy = f (x, y)dy = 2x.
−∞ 0

d
This is consistent with (c) because dx
(x2 ) = 2x.
(e) We first compute fY (y) for 0 ≤ y ≤ 1 as
Z 1
6
fY (y) = f (x, y)dx = y(y + 1).
0 5

5
Since f (x, y) = fX (x)fY (y), we conclude that X and Y are independent.

Problem 17. (a) The marginal probability PY (1) = 1/2

⇒ P (X = 0, Y = 1) = P (X = 2, Y = 1) = 0.
Now each column has one empty entry. This can be computed by making the column
add up to the given marginal probability.
Y \X 0 1 2 PY
-1 1/6 1/6 1/6 1/2
1 0 1/2 0 1/2
PX 1/6 2/3 1/6 1
(b) No, X and Y are not independent.
For example, P (X = 0, Y = 1) = 0 = 1/12 = P (X = 0) · P (Y = 1).

Problem 18. For shorthand, let P (X = a, Y = b) = p(a, b).

(a) P (X = Y ) = p(1, 1) + p(2, 2) + p(3, 3) + p(4, 4) = 34/136.
(b) P (X + Y = 5) = p(1, 4) + p(2, 3) + p(3, 2) + p(4, 1) = 34/136.
(c) P (1 < X ≤ 3, 1 < Y ≤ 3) = sum of middle 4 probabilities in table = 34/136.
(d) {1, 4} × {1, 4} = {(1, 1), (1, 4), (4, 1), (4, 4) ⇒ prob. = 34/136.

Y \X 0 1 PY
X and Y are independent, so the table is computed from 0 1/8 1/8 1/4
Problem 19. (a) the product of the known marginal probabilities. Since 1 1/4 1/4 1/2
they are independent, Cov(X, Y ) = 0. 2 1/8 1/8 1/4
PX 1/2 1/2 1

(b) The sample space is Ω = {HHH, HHT, HTH, HTT, THH, THT, TTH, TTT}.
P (X = 0, Z = 0) = P ({T T H, T T T }) = 1/4.
P (X = 0, Z = 1) = P ({T HH, T HT }) = 1/4. Z\X 0 1 PZ
P (X = 0, Z = 2) = 0. 0 1/4 0 1/4
P (X = 1, Z = 0) = 0. 1 1/4 1/4 1/2
P (X = 1, Z = 1) = P ({HT H, HT T }) = 1/4. 2 0 1/4 1/4
PX 1/2 1/2 1
P (X = 1, Z = 2) = P ({HHH, HHT }) = 1/4.

Cov(X, Z) = E(XZ) − E(X)E(Z).

E(X) = 1/2, E(Z) = 1, E(XZ) = xi yj p(xi , yj ) = 3/4.
⇒ Cov(X, Z) = 3/4 − 1/2 = 1/4.

Z a Z b
Problem 20. (a) F (a, b) = P (X ≤ a, Y ≤ b) = (x + y) dy dx.
0 0

6
 b a
y2 b2 x2 b2 a2 b + ab2
Inner integral: xy + = xb + . Outer integral: b+ x = .
2 0 2 2 2 0 2
2 2
x y + xy
So F (x, y) = and F (1, 1) = 1.
2
Z 1 Z 1 1
y2 1
(b) fX (x) = f (x, y) dy = (x + y) dy = xy + = x+ .
0 0 2 0 2
By symmetry, fY (y) = y + 1/2.
(c) To see if they are independent we check if the joint density is the product of
the marginal densities.
f (x, y) = x + y, fX (x) · fY (y) = (x + 1/2)(y + 1/2).
Since these are not equal, X and Y are not independent.
Z 1Z 1 Z 1" 2 1
# Z 1
2 y x 7
(d) E(X) = x(x + y) dy dx = x y+x dx = x2 + dx = .
0 0 0 2 0 0 2 12
Z 1 Z 1
(Or, using (b), E(X) = xfX (x) dx = x(x + 1/2) dx = 7/12.)
0 0
By symmetry E(Y ) = 7/12.
Z 1Z 1
2 2 5
E(X + Y ) = (x2 + y 2 )(x + y) dy dx = .
0 0 6
Z 1Z 1
1
E(XY ) = xy(x + y) dy dx = .
0 0 3
1 49 1
Cov(X, Y ) = E(XY ) − E(X)E(Y ) = − = − .
3 144 144
Problem 21.
Standardize:
! 1P
Xi − µ

X
n 30/n − µ
P Xi < 30 =P √ < √
i
σ/ n σ/ n
 
30/100 − 1/5
≈P Z< (by the central limit theorem)
1/30
= P (Z < 3)
= 1 − .0013 = .9987 (from the table)

X1 + . . . + X144
Problem 22. Let X = ⇒ E(X) = 2, and σX = 2/12 = 1/6.
√ 144
( n = 12)
 
264

A chain of algebra gives P (X1 + . . . + X144 ) + P X > = P X > 1.8333 .
144

7
    
X −2 1.8333 − 2 X −2
Standardization gives P (X > 1.8333) = P > =P > −1.0
1/6 1/6 1/6
 
X −2
Now, the Central limit theorem says P > −1.0 ≈ P (Z > −1) = .84
1/6

Problem 23. Let Xj be the IQ of a randomly selected person. We are given

E(Xj ) = 100 and σXj = 15.
Let X be the average
√ of the IQ’s of 100 randomly selected people. We have (X) = 100
and σX = 15/ 100 = 1.5.
The problem asks for P (X > 115). Standardizing we get P (X > 115) ≈ P (Z > 10).
This is effectively 0.

Problem 24. Data mean and variance x̄ = 65, s2 = 35.778. The number of
degrees of freedom is 9. We look up t9,.025 = 2.262 in the t-table The 95% confidence
interval is
t9,.025 s t9,.025 s √ √
x̄ − √ , x̄ + √ = 65 − 2.262 3.5778, 65 + 2.262 3.5778 = [60.721, 69.279]
n n

Problem 25. Suppose we have taken data x1 , . . . , xn with mean x̄. Remember in
these probabilities µ is a given (fixed) hypothesis.
   √ 
|x̄ − µ| .5 .5 n
P (|x̄−µ| ≤ .5 | µ) = .95 ⇔ P √ < √ | µ = .95 ⇔ P |Z| < = .95.
σ/ n σ/ n 5
√
.5 n
Using the table, we have precisely that = 1.96. So, n = (19.6)2 = 384. .
5
√
If we use our rule of thumb that the .95 interval is 2σ we have n/10 = 2 ⇒ n = 400.
√
Problem 26. The rule-of-thumb is that a 95% confidence interval is x̄ ± 1/ n.
To be within 1% we need
1
√ = .01 ⇒ n = 10000.
n
Using z.025 = 1.96 instead the 95% confidence interval is
z.025
x̄ ± √ .
2 n
To be within 1% we need
z.025
√ = .01 ⇒ n = 9604.
2 n

8
Note, we are using the standard Bernoulli approximation σ ≤ 1/2.

Problem 27. The 90% confidence interval is

z.05 1.64
x± √ =x±
2 n 40
1.64
We want x − 40
> .5, that is x > .541.
number preferring A
So 400
> .541. So,

number preferring A > 216.4

Problem 28. A 95% confidence means about 5% = 1/20 will be wrong. You’d
expect about 2 to be wrong.
With a probability p = .05 of being wrong, the number wrong follows
l a Binomial(40, p)
distribution. This has expected value 2, and standard deviation 40(.05)(.95) = 1.38.
10 wrong is (10-2)/1.38 = 5.8 standard deviations from the mean. This would be sur­
prising.

Problem 29. We have n = 27 and s2 = 5.86. If we fix a hypothesis for σ 2 we

know
(n − 1)s2
2
∼ χ2n−1
σ
We used R to find the critical values. (Or use the χ2 table.)
c025 = qchisq(.975,26) = 41.923
c975 = qchisq(.025,26) = 13.844
The 95% confidence interval for σ 2 is
(n − 1) · s2 (n − 1) · s2
   
26 · 5.86 26 · 5.86
, = , = [3.6343, 11.0056]
c.025 c.975 41.923 13.844

We can take square roots to find the 95% confidence interval for σ

[1.9064, 3.3175]

Problem 30. (a) The model is yi = a + bxi + εi , where εi is random error.

We assume the errors are independent with mean 0 and the same variance for each i
(homoscedastic).
The total error squared is
1
E2 = (yi − a − bxi )2 = (1 − a − b)2 + (1 − a − 2b)2 + (3 − a − 3b)2

9
The least squares fit is given by the values of a and b which minimize E 2 . We solve
for them by setting the partial derivatives of E 2 with respect to a and b to 0. In R
we found that a = 1.0, b = 0.5
(b) This is similar to part (a). The model is

yi = axi,1 + bxi,2 + c + εi

where the errors εi are independent with mean 0 and the same variance for each i
(homoscedastic).
The total error squared is
1
E2 = (yi − axi,1 − bxi,2 − c)2 = (3 − a − 2b − c)2 + (5 − 2a − 3b − c)2 + (1 − 3a − c)2

The least squares fit is given by the values of a, b and c which minimize E 2 . We solve
for them by setting the partial derivatives of E 2 with respect to a, b and c to 0. In R
we found that a = 0.5, b = 1.5, c = −0.5

10
MIT OpenCourseWare
http://ocw.mit.edu

18.05 Introduction to Probability and Statistics

Spring 2014

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.