Stelsels en Seine / Systems and Signals 344 Tutorial 5 Memo
Question 1
The CDF for X is
(a)
P (X > 0.75) = 1 − P (X ≤ 0.75)
= 1 − FX (0.75)
= 1 − 0.875
= 0.125
(b)
P (−0.5 < X ≤ 0.5) = FX (0.5) − FX (−0.5)
3 1
= −
4 4
1
=
2
(c)
P (|X| ≤ 0.5) = P (−0.5 ≤ X ≤ 0.5)
= P (−0.5 < X ≤ 0.5) + P (X = −0.5)
= 0.5 + 0
= 0.5
(d)
a+1
P (X ≤ a) = FX (a) = = 0.6
2
a = 0.2
Stelsels en Seine / Systems and Signals 344 Tutorial 5 Memo
Question 2
(a)
P (X ≤ 5) = FX (5)
5−8
=F √
4
= Φ(−1.5)
= 1 − Φ(1.5)
= 0.0668
(b)
P (4 < X ≤ 6) = FX (6) − FX (4)
6−8 4−8
=Φ √ −Φ √
4 4
= Φ(−1) − Φ(−2)
= 1 − Φ(1) − (1 − Φ(2))
= 0.1587 − 0.02275
= 0.13595
(c)
P (X > 9) = 1 − FX (9)
9−8
=1−Φ √
4
= 1 − Φ(0.5)
= 0.3085
Question 3
We need to find the distribution parameters of T . With that we will be able to find P (T > 0). We have the
standard deviation as √
σT = 100 = 10
and we can find the mean using
−12 − µT
P (T > −12) = 1 − P (T ≤ 12) = 1 − Φ = 0.5
10
Using our handy table for z-values, we find that for Φ(z) = 0.5 to be true we need z = 0. So we can find that
−12 − µT
=0
10
∴ µT = −12
Note that you could directly have assumed −12 to be the mean, because the CDF of a Gaussian distribution is 0.5
at the mean. With these parameters we can find the probability of T being above freezing:
P (T > 0) = 1 − P (T ≤ 0)
0 − (−12)
=1−Φ
10
= 1 − Φ(1.2)
= 0.1151
Stelsels en Seine / Systems and Signals 344 Tutorial 5 Memo
Question 4
(a) The exponent and coefficient in the expression for the probability density function of an exponential RV are
both simply the rate parameter, as used to describe the mean number of outcomes per unit time in the case
of a Poisson RV. This is because, if the number of outcomes per unit time is described by a Poisson random
variable, then the time between events is modeled by an exponential random variable. The mean number of
times per hour that the computer crashes is the reciprocal of the mean time between computer crashes, which
is λ1 . Given the CDF, the corresponding PDF is
5e−5x
0≤x
fX (x) =
0 otherwise
Comparing the form of the PDF to the table of PDFs of continuous random variables, the mean time between
crashes is 15 . The reciprocal is 5. Therefore the mean number of times per hour that the device crashes is 5
times.
(b) The probability of the computer not crashing in ten minutes can be calculated as
1 1 1
P X> =1−P X≤ = 1 − FX = 0.435
6 6 6
Question 5
4
We are given that m1 = 3 and we know that
2
σX = m2 − m21 = E[X 2 ] − E[X]2 ,
so we only need to calculate m2 :
Z ∞
m2 = E[X 2 ] = x2 fX (x)dx
−∞
Z 4
1 1
= x2
− x+ dx
0 8 2
1 4 3 1 4 2
Z Z
=− x dx + x dx
8 0 2 0
4 4
1 1 4 1 1 3
=− x + x
8 4 0 2 3 0
2
=2
3
We can calculate the variance as
2
σX = m2 − m21
2
2 4
=2 −
3 3
8
=
9
Stelsels en Seine / Systems and Signals 344 Tutorial 5 Memo
Question 6
(a) The area under a PDF must add up to 1, so
Z ∞
fX (x)dx = 1
−∞
20 Z 30 Z 60
−a
Z
a
= x − a dx + adx + x + 2a dx
10 10 20 30 30
20 30 60
a −a 2
= x2 − ax + ax + x + 2ax
2 · 10 10 20 2 · 30 30
= 5a + 10a + 15a = 1
1
∴a=
30
Since the integral simply computes the area of the function, if you used geometric arguments to sum up the
areas, ie, sum the area of the first triangle with the area of the rectangle plus the area of the second triangle,
you would still be correct (the answer will be identical).
Z ∞
fX (x)dx = 1
−∞
1 1
= b1 h + b2 h + b3 h = 1
2 2
1 1
= · 10 · a + 10 · a + · 30 · a = 1
2 2
1
∴a=
30
(b)
Z 25
P (X ≤ 25) = fX (x)dx
−∞
Z 20 Z 25
a
= x − a dx + adx
10 10 20
= 5a + 5a
= 10a
1
=
3
or from the geometric argument:
1 1 1
P (X ≤ 25) = b1 h + b2 h = · 10 · a + 5 · a =
2 2 3
(c) The cumulative distribution function is
Stelsels en Seine / Systems and Signals 344 Tutorial 5 Memo
Question 7
NOTE: We have not covered this in class, but it is a good exercise to do. We will look at how to solve these
problems in Lecture 18.
1
(a) We first need to calculate r. We can use 2πr = 1 to determine r = 2π . The CDF of Y can be expressed as
X
FY (y) = P (Y ≤ y) = P ≤ y = P (X ≤ 4πy) = FX (4πy)
4π
The random variable X is uniformly distributed, so the CDF of X is simply FX (x) = x for 0 ≤ x ≤ 1
(remember, the circumference is 1). This means that for 0 ≤ x < 1 the CDF of X is FX (x) = x. Therefore
the CDF of Y is
0
for y < 0,
1
FY (y) = 4πy for 0 ≤ y < 4π
1
1 for y ≥ 4π
(b) We can get the PDF of Y by taking the derivative of the CDF:
(
1
4π for 0 ≤ y < 4π
fY (y) =
0 otherwise
(c) The expected value of Y is
Z 1
4π 1
E[Y ] = 4πy dy =
0 8π
