DS 300
4/11/2017
Linear regression
Conditions for linear regression
1. The error
is a random variable with the mean of zero
2. The variance of
, denoted by
is the same for all variables
DS 300
1/24/17
FrequencyHow many?
Relative FrequencyWhat fraction of observations are in a class
Cumulative FrequencyHow many observations have a value less than or equal to a number
Cumulative Relative FrequencyWhat fraction of observations have a value
DS 300
2/9
Marginal probability- you find them at the margins of the table
Independence
Two events are independent if
Rolling a fair dice produces independent events
0.313947
0.686053
0.309634
0.576706
0.423294
0.773373
3.541457
-2.541457 -2.541457
sample mean
population mean
sample standard dev
number in sample
test statistics for use with t
number of degrees of f
P(t>Test statistics in cell b6)
252.45
45
4.6
36
270.
DS 300
1/31/17
Single Variable Statistics
Understand the purpose of measure of location
Understand the purpose of measures of variation
Know when to use which measure
Two Variable Statistics
Review Scatter Plots
Understand the purpose of the measure of as
DS 300
2/14/17
Discrete Random Variables (not on exam 1)
-count how many, limited number of possible types of values
-how many cats live in your household?
Continuous variables
-measure how much
-if we had a precise enough measuring device could be any po
DS 300
3/23/2017
Topic 9
Increase in confidence = increase in margin of error
Point estimate
Critical value
margin of error
*
Std error = magnified
Point estimator
*Bigger degrees of freedom = looks more like standard normal
t.dist.rt for upper tail proba
DS 300
2/7
Classical method
Assigning probabilities based on the assumption of equally likely outcomes
Relative Frequency Method
Assigning probabilities based on experimentation or historical data
Subjective Method
Assigning probabilities based on judgmen
true= cumulative probability is desired
always true for fourth input
false= if the height of the curve is desired
0.1586552539
0.8413447461
standard dev = 1, mean= 0
0.6914624613
0.308538
0.5998593133
-1.2815515655
0
1
2
3
0.3
0.5
0.15
0.05
0.9025
0.0025
Dishwasher Tablets Sample Exam Question
You work for a dishwasher tablet manufacturer. You have been concerned for some time about the
variability of the weight of one dishwasher tablet. Ideally, dishwasher tablets should weigh exactly
20.1g, however, any
LisaTotal
FauberMichigan
Statistical
residents obtained a Bachelors Degree (484) or
Report
1
advanced
degrees (315) is 799 out of the population tested (27.20%).
Total population is 344,816 thousand civilian noninstitutional
Americans (bls.gov). 72,528 t
Quality of Kitchens in 2010 with Normal Conditions of Sale
140
119
120
95
100
Number of Kitchens
80
60
40
20
0
6
7
Ex
Fa
1
Gd
Po
TA
Kitchen Quality
Lisa Fauber Statistics Report 2
Yr Sold
Sale Condition
Kitchen quality
Ex
Fa
Gd
Po
TA
Grand Total
2010
Norm
DS 300
3/21/17
Coefficient of variation: bigger percentage means more variable
-readahead attached to homework, fill in readahead so the symbols and formulas are straight
-excel quiz on t.inverse (before the class)
95% confident that in 2014, in the USA,
DS 300
3/28/17
s
Significance test:
Need statement about how the way the world is supposed to be
Need statement about how the way things are
Hypothesis testing can be used to determine whether a statement about the value of a population
parameter should o
DS 300
2/2
Topic 4 Descriptive Measures
Resistive Measures- picks up whats at the beginning and the end
Scatter diagram- x-y coordinate plot
Sample covariance
-covarience will add it all up and youll get a sum
-Divide it up by the number of this (in a sam
DS 300
2/23/17
NORM.S.DIST function for z-score normal table
Continuous random variable can assume any value in an interval on the real line or in a collection of
intervals
-is not possible to talk about the probability of the random variable assuming a a
DS 300
1/26/17
Simpsons ParadoxData in 2 or more crosstabulations are often aggregated to produce a summary crosstabulation
-We must be careful in drawing conclusions about the relationship between the two variables in the
aggregated
-lurking variable (in
DS 300
1/12/2017
Sample
N=100
Categorical Variables-have values which indicate a category/type
-names things
-sometimes qualitative
-do not overlap (Mutually exclusive)
-contain all values of the variables (collectively exhausted)
Ex: Drop down bars on we
Dora Furtos
Assignment 10
21 November 2016
13.7
a.) Scatter plot shows that there is a positive linear relationship between the two variables
b.) The regression coefficients for b0 s 0.75 and b1 is 0.5.
c.) The coefficient value for Plate Gap is 0.5. This
Part A)
Figure 1: How many bedrooms are there in Midwestern homes sold in 2013?
Number of Bedrooms in Midwest Homes Sold in 2013
0.7
0.6
0.5
0.4
Percentage of Homes in Sample 0.3
0.2
0.1
0
1-3
4+
Number of Bedrooms
Source: Statistical Report Spreadsheet
I
.
1) Are Michigan residents more or less likely than Americans to have a High school graduates, no college?
Less likely
America 29.5% (62,578/212,132), Michigan 25.9% (463/1785)
2) How easy will it be to actually hire employees in Michigan?
Hiring empl
12-20
(a)
(b) McDonalds uses more debt with a much higher Debt-to-equity ratio. Although their Debtto-equity ratio is still below the industry average.
(c) McDonalds and Wendys both exceed the industry averages, except in Debt-to-equity. This
is also good
Assignment 2
Lisa Gilmore
Alexis Guerra
William Losel
Hassan Hammoud
Mohamed Ali Taha
Gilmore-Guerra-Hammoud-Loesel-Taha Group
Page 56 Problem 2.27
C)
D) The reason for the vital few complaints are that airbags and seatbelt issues
are usually only discove
Port
Baltimore
Baton Rouge
Beaumont
Corpus Christi
Duluth-Superior
Houston
Huntington
Lake Charles
Long Beach
Los Angeles
Mobile
New Orleans
New York
Norfolk Harbor
Pascagoula
Philadelphia
Pittsburgh
Plaquemines
Port Arthur
Savannah
South Louisiana
St. Lo