Math 183 Quiz Solutions - Expected Values, Confidence Intervals, Binomial Dist. (Fall '06), Quizzes of Data Analysis & Statistical Methods

The solutions to quiz 3 for math 183, a college-level mathematics course taken in the fall of 2006. The solutions cover topics such as expected values of continuous random variables, binomial distribution, and confidence intervals. Students can use this document to check their understanding of these concepts and to study for exams.

Typology: Quizzes

Pre 2010

Uploaded on 03/28/2010

koofers-user-k5v
koofers-user-k5v 🇺🇸

10 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Math 183
Fall 2006
Quiz 3 Solutions
1a. The expected value of a continuous random variable is given by
E[Y] = ZyfY(y)dy
where the integral is over the domain of fY(y). So for part awe have
E[Y] = Z1
0
y·3(1 y)2dy
=Z1
0
(3y36y2+ 3y)dy
=1
4
1b. Here we have
E[Y] = Z
−∞
fY(y)dy
= 0 + Z1
0
y·3
4dy + 0 + Z3
2
y·1
4dy + 0
=3
8+5
8
= 1
2. The probability that a single observation of yfalls in the interval 1
2,1is
given by integrating the pdf:
P1
2Y1=Z1
1
2
3y2dy
=y3
1
1
2
=7
8
1
pf3

Partial preview of the text

Download Math 183 Quiz Solutions - Expected Values, Confidence Intervals, Binomial Dist. (Fall '06) and more Quizzes Data Analysis & Statistical Methods in PDF only on Docsity!

Math 183

Fall 2006

Quiz 3 Solutions

1a. The expected value of a continuous random variable is given by

E[Y ] =

yfY (y)dy

where the integral is over the domain of fY (y). So for part a we have

E[Y ] =

0

y · 3(1 − y)^2 dy

0

(3y^3 − 6 y^2 + 3y)dy

1b. Here we have

E[Y ] =

−∞

fY (y)dy

0

y ·

dy + 0 +

2

y ·

dy + 0

  1. The probability that a single observation of y falls in the interval

[ 1

2 ,^1

]

is given by integrating the pdf:

P

[

≤ Y ≤ 1

]

(^12) 3 y^2 dy

= y^3

1 (^12)

=

We now let X be the number of observations in this interval out of 15 samples. This is equivalent to flipping a biased coin with probability 78 of heads 15 times, and counting the number of heads. In other words, X is binomially distributed with n = 15 and p = 78. The expected value of X can be computed from the formula for the expected value of a binomial random variable:

E[X] = np = 15

3a. The “confidence” of an interval is the area over the corresponding region under the normal curve. So in part a, the confidence is:

ϕ(2.33) − ϕ(− 1 .64) ≈ 0. 99 − 0. 05 ≈ 0. 94

3b. Since the left endpoint is −∞, we just have the area to the left of the right endpoint.

ϕ(2.05) ≈ 0. 98

3c. Since y = y + 0 · σ/

n, we have

ϕ(0) − ϕ(− 1 .64) ≈ 0. 50 − 0. 05 ≈ 0. 45

  1. For a binomial random variable, we can derive from the central limit theorem that for large n

P

[

X − z α 2

σ √ n

≤ p ≤ X + z α 2

σ √ n

]

≈ 1 − α

We don’t know σ, but since n is large

σ ≈

X(1 − X)

We wish to construct a 96% confidence interval, so α = 0.04 and

z α 2 = z 0. 02 ≈ 2. 05

Thus our confidence interval is [

  1. 55 −

]

≈ [0. 45 , 0 .65]

  1. For a 98% confidence interval, α = .02 and

z α 2 = z 0. 01 ≈ 2. 33