12_13_quant stat

# 12_13_quant stat - Review of Quantitative Statistics CS 378...

This preview shows pages 1–13. Sign up to view the full content.

1 Review of Quantitative Statistics CS 378 Lecture 6

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 Two Branches • Descriptive Statistics How can we report or display data? • Inferencial Statistics What meaning can we obtain from the data?
3 Descriptive Statistics Example Data 3, 5, 3, 4, 2, 5, 10, 2, 1, 5, 8 • Measure of Central Tendancy Mean X = 4.36 œ ¸ ± 3œ" 8 X n1 1 48 i (the center of mass or the balance point) Median X 4 ~ X n odd n even œ H ÐÑ 8" # Ð " Ñ 88 ## XX 2 + where X is ith smallest X value Ð3Ñ (the center of probability) Mode MAX(X ) 5 œ i

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4 Descriptive Statistics Example Data 3, 5, 3, 4, 2, 5, 10, 2, 1, 5, 8 • Variability Variance 6.6 œ œ ±± 3œ" 3œ" 88 ## ÐÑ 3œ" 8 # (X X (X ) nn ii X) i n Ñ ± (division by n 1 for inference) Range Max(X ) Min(X ) 9 œ 3 i Inter Quartile Range 75th %-ile 25th %-ile • Other Measures Skewness Kurtosis
5 Descriptive Statistics Displaying Data Single Group 8.7 10.1 10.6 8.7 11.1 9.7 10.3 10.9 9.0 8.4 9.2 9.4 11.2 10.8 9.9 7.5 9.4 10.1 12.2 10.0 8.7 8.6 9.4 9.1 10.2 9.8 11.7 11.2 11.0 11.0 10.0 10.4 9.9 10.3 9.9 10.9 10.3 8.8 7.6 9.7 8.4 10.1 10.8 9.6 10.5 9.9 10.2 10.3 9.2 10.1 9.7 10.2 11.4 12.0 7.9 10.6 8.4 8.5 9.1 8.8 11.0 9.3 9.6 9.3 9.9 10.5 10.5 10.0 10.6 10.1 11.1 8.4 10.1 11.8 10.8 9.9 8.6 10.2 10.1 9.7 10.0 9.8 10.3 9.0 10.1 9.4 10.1 11.3 8.7 11.1 10.6 10.6 10.2 8.9 11.0 10.7 9.0 9.4 9.6 10.5

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
6 Histograms Cumulative Frequency Plots
7 Descriptive Statistics Displaying Data Multiple Group Observations I n d e p V a r 123456T o t a l s A v e r a g e 5 7 8 15 11 9 10 60 10.00 10 12 17 13 18 19 15 94 16.54 15 14 18 19 17 16 18 102 12.00 20 19 25 22 23 18 20 127 21.17 383 15.96 Box Plots

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
8 Tier Plots
9 Inferential Statistics Random Sample only random (not systematic or Ê induced) error is involved Random Sampling Methods Simple use a randomization technique (eg. random Ê number generator) to select sample Stratified Sampling to account for a variables representation in the population (eg income level) 1. break up population into strata (groupings) of that variable (eg <\$10,000, [\$10,000, \$50,000), etc) 2. keep starta %-age in sample the same as in population but randomly sample within each starta Real Life Issue have to work with what you have--be Ê honest & document.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
10 Inferential Statistics - Concepts • Parametric vs Nonparametric Parametric - X has some distributional form f (i.e. i normal) but we do no know the parameters, use data to estimate the parameters. Non parametric - no distributional form is assumed Parametric is easier and more efficeint but if the distributional assumption is invalid results are meaningless
11 Inferential Statistics - Concepts Point Estimation Basics Given an unknown parameter and sample X , . .., X ) " n what are some estimators for ? They are functions @) of the random sample i.e. (X) 3, or max(X) or , . ..... ~~ @ œ ± 3œ" 8 3 X n an estimator is a random variable woth a probability Ê distribution and an estimate is a realization of that random variable! What is a good estimator? Properties unbiassedness E[ (X)] and inimum variance ~ œ

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
12 Inferential Statistics - Concepts Interval Estimation Basics (Confidence Interval) Procedure 1. find some function h( (X), ) of the estimator with ~ @) a known distribution, 2. find percentage points such P r { h ( ( X ) , ) } 1 ~ )@ ) ) α αα /2 1 /2 ŸŸ œ s o l v e f o r . ) Examples is N(0,1) X /n .
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 54

12_13_quant stat - Review of Quantitative Statistics CS 378...

This preview shows document pages 1 - 13. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online