Review for Final Examination 1F03 Statistics
1. There theoretically are an infinite number of possible values between any two adjacent
scale values with a
variable.
a. nominal
b. continuous
c. discrete
d. measured
2. A _ is any measurable characteristic t

Review for Final Examination 1F03 Statistics - ANSWERS
1. There theoretically are an infinite number of possible values between any two adjacent
scale values with a
variable.
a. nominal
b. continuous
c. discrete
d. measured
2. A variable is any measurable

Practice Midterm Questions Set C. These questions have not been reviewed and may
contain errors. They are not meant to be representative of the difficulty, length nor
distribution of questions of the actual midterm. Remember the test is multiple choice
bu

HTH SCI 2A03 Midterm Version 0
Circle your tutorial section
Date:
Thursday, March 12, 2015
Duration: 50 minutes (17301820)
Instructor: Mateen Shaikh
Fill in the following info
Username
(MacID)
Name
Student
Number
T03
Mon 10:30 AM
Zahra
T08
Mon 11:30 AM
Za

Practice Final A
These questions have not been reviewed and may contain errors. Though not all of
these questions are, the actual midterm is entirely multiple choice.
Consider an investigation of the dimensions of a petal of the Versicolor Iris (its prett

1. A university dean of students wishes to estimate with 95% confidence the average
number of hours students spend doing homework per week. It has been estimated
that the population standard deviation is about 6.2 hours. How large a sample
must be selecte

Practice Exam Questions
1. Consider the data set of test marks that is summarized in the output below
Mark grouping
020-029
030-039
040-049
050-059
060-069
070-079
N= 16
Count
?
?
?
?
?
?
Percent
18.75
18.75
12.50
12.50
18.75
18.75
Fill in the missing cou

Practice Final ASaturday 11th April, 2015,11:52
These questions have not been reviewed and may contain errors. Though not all of
these questions are, the actual midterm is entirely multiple choice.
Consider an investigation of the dimensions of a petal of

Practice Midterm Questions Set A. These questions have not been reviewed and may
contain errors. They are not meant to be representative of the difficulty, length nor
distribution of questions of the actual midterm. Remember the test is multiple choice
bu

Practice Final B
These questions have not been reviewed and may contain errors. Though not all of
these questions are, the actual midterm is entirely multiple choice.
Consider an investigation of the girth of a tree against the height of the tree. Let X
r

Practice Final B
These questions have not been reviewed and may contain errors. Though not all of
these questions are, the actual midterm is entirely multiple choice.
Consider an investigation of the girth of a tree against the height of the tree. Let X
r

Practice Midterm Questions Set B. These questions have not been reviewed and may
contain errors. They are not meant to be representative of the difficulty, length nor
distribution of questions of the actual midterm. Remember the test is multiple choice
bu

Linh and Arunimas
Solutions to the Statistics
Practice Exam!
Linh and Arunimas Solutions to the Statistics Practice Exam
1. A university dean of students wishes to estimate with 95% condence the
average number of hours students spend doing homework per w

7
5
4
3
2
1
Length in cm
6
setosa
versicolor
virginica
0.0
0.5
1.0
1.5
2.0
2.5
Width in cm
1
When we have two variables, more than just mean and variances happen. Centers
are still centers, and variances within each variable still happen, but now we can
a

We will briey talk about terms that usually goes at the beginning of the
course, but the dierence between the terms are actually used now.
Denitions
A unit is an entity which provides a single set of measurements.
A population is a collection of units, an

Intervals
Intervals - centered around how often do they contain a random X ?
Shift the same interval to be centered around X , how often will it contain ?
The same because the interval is the same width
So infer something about what we think could be by c

Denition
Given the discrete random variable X cfw_x1 , x2 , . . . , xn with distribution
P(X ), the Expected Value of some function of X , say g (X ), is given by
n
E(g (X ) =
g (xi )P(xi )
i=1
Denition
Given the discrete random variable X cfw_x1 , x2 ,

A paediatrician will see 8 girls and 7 boys a given day, and will give a cat sticker to 5 of them.
Letting X be the number of girls that get cat stickers, nd
Example
A paediatrician will see 8 girls and 7 boys a given day, and will give a cat sticker to 5

Descriptive Statistics and Plots
HTH 2A03
January 15, 2015
R
The software I recommend for stats in this course is R
Available from the r website http:/cran.r-project.org/
CRAN provides an IDE for Windows and Mac but I like the
IDE R studio from http:/www.

Intervals for Single observations
Example
Consider the example of full term infant birth weights by mothers that had
gestational diabetes (GDM). The average weight in a particular population is
3.3kg with a standard deviation of 0.6kg. If the distribution

Proportions
Example
A radiologist examines many images and over the course of any given month, an
average of 43% of ordered cultures indicate a tumour. What is the probability
that among 10 independent images, at most 2 indicate a tumour? At most 7
indica

Descriptive Statistics and Plots
HTH 2A03
Lots of terminology
Denitions
An experiment is some process that yields an outcome.
Denition
If dierent outcomes are possible due to an element of randomness, the
process is stochastic.
If, however, the outcome wi

Intervals for Single observations
Example
Consider the example of full term infant birth weights by mothers that had
gestational diabetes (GDM). The average weight in a particular population is
3.3kg with a standard deviation of 0.6kg. If the distribution

More tests
Another distribution we can take advantage of using the CLT is the 2
distribution, dened as the distribution when calculating the sum of squared
random normal distributions:
Denition
2
2
2
2
Suppose Y = Z1 + Z2 + Z3 + + Zn where the Z s are ind

Thus far the distributions we have considered have been discretely, largely
count data. Variables, like measurements, however, can be continuous
variables. They are handled slightly dierently.
Consider the example where values are equally likely
x
0
1
P(x

2. t-test
Why use t-test?
Problems with the z-test: comparing to a population mean, which has no
error, so the only error is occurring from the sample mean, whereas when
comparing means of two samples, you have error in both groups, thus
need new test
Whe

5. Multiple Regression
o
Assumptions are the same as for simple regression
o
Multiple regression: linear relationship between one continuous dependent variable and more than one
independent variables
Ex) Comparing FEV1 (dependent) to age, sex, height (ind

ANOVA Tests
Analysis of variance (ANOVA): testing differences in the means of more
than two groups
Assumptions:
Variances are the same (homoscedasticity)
Normal distribution
Each of the observations are independent from one
another
Hypotheses:
Null hypoth

Regression: Determining
the form of a relationship
(Part II)
Multiple linear regression
1
Assumptions of Regression
Model
Existence: T h e re la tio ns h ip b e twe e nd e p e nd e nta nd
ind e p e nd e ntva ria b le s e xis t
Linearity: T h e re la tio