Professional Documents
Culture Documents
Nature of Statistics
Nature of Statistics
Nature of Statistics
1 Two kinds of statistics Statistics Body of principles and methods concerned with Two basic areas: 1. Descriptive Statistics Deals with methods of Ex. 2. Inferential Statistics Body of methods for drawing conclusions about a
- Parameter
- Sample
- Statistics
Classifying Statistical Studies Descriptive Describes & summarizes data. ( ). Ex. The voting results for A bank manager determines the
Inferential
Ex. A consumer group wants information about Biologist interested on body size of
1.3 Simple Random Sampling (SRS) Each item of a population is Two types: 1. SRS with replacement
Here: Assume SRS without replacement. How large should be a sample? Often surprisingly small. Pre-election polls: Population Sample Error
9
1.4 Other Sampling Designs 1.5 Experimental Design Observational Studies characteristics and take measurements. Design Experiment then observe characteristics and take measurements.
10
DESCRIPTIVE STATISTICS (Weiss, Chapters 2 & 3) 2.1 Variables and Data Variable A characteristic that Data Information obtained by Two types of data Quantitative Response Ex: Qualitative Response Ex:
11
Quantitative Data Discrete Whose possible values form a finite (or countable infinite) set of numbers. Ex. Continuous Whose possible values form some interval of numbers. Ex. Important: The statistical technique employed depends on the type of data. Qualitative Quantitative -
12
Summary: Variable
13
2.2 Grouping Data The Bureau of Economic Analysis gathers information on the length of stay in Europe by US travelers. A sample of 36 US residents yield the following data: 41 5 3 21 3 1 16 31 32 44 13 11 6 20 2 1 15 14 21 27 48 56 10 12 1 17 8 5 18 64 21 10 12 12 3 10
Group data using classes of equal width starting with the class 0 - (12). 0 - (12): 0 inclusive up to 12 exclusive - up to and not including.
14
Length of Stay 0 - (12) 36 - (48) 48 - (60) 60 - (72) Guidelines for grouping data:
Frequency
2 2 1
1. # classes small enough to provide & large enough to 2. Each observation must 3. Whenever feasible, all classes
15
Frequency & Relative-Frequency Frequency The number of Relative Frequency expressed as a decimal. Hence Relative Frequency = where n: fi : Midpoint
16
Freq. 15 13 3 2 2 1
Midpoint
42 54 66
Single-Value Grouping Sometimes more appropriate to use classes that each represents a single possible value: - Discrete data with few observations, - Qualitative data.
17
Example 2.8, pg 50 Students in an introductory stats class were asked for their political party affiliation: - Democratic (D) - Republican (R) - Other (O) Responses: D D D D O R O R O R O R O D D R D D D R R O R D R R O R R R R R O O R R D R D D
Party
Frequency
Relative Frequency
18
2.3 Graphs and Charts Histograms Ex. US travelers in Europe Length of Stay 0 12 24 36 48 60 - (12) - (24) - (36) - (48) - (60) - (72) Freq. 15 13 3 2 2 1 Relat. Freq. 0.417 0.361 0.083 0.056 0.056 0.028
19
1.
2.
Relative-frequency Histogram: Plot Classes Relat. Freq. Note: Freq. and rel. freq. are both represented by height is the
20
Frequency
2 42
2 54
1 66
18
30
21
Relative Frequency
0.083
0.056
0.056
0.028 66
18
30
42
54
22
Cumulative Frequency Distributions Usually cumulate for Number of observations Ex. US travelers: < < 36 days 36 days < class end point
23
Ogive (pg 65) Plot relative cumulative frequencies against upper end points of classes. Ex. US travelers
Freq 0 15 13 3 2 2 1
33 35 36
24
OGIVE - US Travelers
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 12 24 36 48 60 72 Length of Stay (days)
Uses of ogives What is the length of stay that 50% of US travelers fall above (below) of? (known as Median.) Read off the ogive: What is the minimum length of stay the top 10% of travelers stay in Europe? Approximately
25
Percentiles
100% P%
Pth percentile
The pth percentile is a number Ex. US travelers First quartile Median Third quartile 10th percentile . th percentile . th percentile . th percentile . 90th percentile .
26