Simple Random Sampling

advertisement

Last lecture summary

• Which measures of variability do you know?

• What are they advantages and disadvantages?

• Empirical rule

Statistical jargon population (census) vs. sample parameter (population) vs. statistic (sample)

Population - parameter

Mean πœ‡

Standard deviation 𝜎

Sample - statistic

Mean π‘₯

Standard deviation s

V ýbΔ›r - statistika

VýbΔ›rový prΕ―mΔ›r π‘₯

VýbΔ›rová smΔ›rodatná odchylka s

Statistical inference

• A statistic is a value calculated from our observed data

(sample).

• A parameter is a value that describes the population.

• We want to be able to generalize what we observe in our data to our population. In order to this, the sample needs to be representative .

• How to select a representative sample? Use randomization .

New stuff

Random sampling

• Simple Random Sampling (SRS) – each possible sample from the population is equally likely to be selected.

• Stratified Sampling – simple random sample from subgroups of the population

• subgroups: gender, age groups, …

• Cluster sampling – divide the population into nonoverlapping groups (clusters), sample is a randomly chosen cluster

• example: population are all students in an area, randomly select schools and create a sample from students of the given school

Simple random sampling

• sampling with replacement (WR)

• výbΔ›r s navrácením

• Generates independent samples

• Two sample values are independent if that what we get on the first one doesn't affect what we get on the second.

• sampling without replacement (WOR)

• výbΔ›r bez navrácení

• Deliberately avoid choosing any member of the population more than once.

• This type of sampling is not independent, however it is more common.

• The error is small as long as

1.

2.

the sample is large the sample size is no more than 10% of population size

Bias

If a sample is not representative, it can introduce bias into our results.

bias – zkreslení, odchylka

• A sample is biased if it differs from the population in a systematic way.

• The Literary Digest poll, 1936, U. S. presidential election

• surveyed 10 mil. people – subscribers

• 2.3 mil. responded predicting (3:2) a Republican candidate to win

• a Democrat candidate won

• What went wrong?

• only wealthy people were surveyed (selection bias) survey was voluntary response (nonresponse bias) – angry people or people who want a change

Bessel’s correction 𝑠 =

π‘₯ 𝑖

− 𝑛 − 1

2 www.udacity.com – Statistics

Sample vs. population SD

• We use sample standard deviation to approximate population paramater σ 𝑠 =

π‘₯ 𝑖

− π‘₯ 𝑛 − 1

2

≈ 𝜎 =

π‘₯ 𝑖

− πœ‡ 𝑛

2

• But don’t get confused with the actual standard deviation of a small dataset.

• For example, let’s have this dataset: 5 2 1 0 7. Do you divide by 𝑛 or by 𝑛 − 1 ?

Bessel's game πœ‡ = 2

8 𝜎 2 =

3

Bessel's game

• An important property of a sample statistic that estimates a population parameter is that if you evaluate the sample statistic for every possible sample and average them all, the average of the sample statistic should equal to the population parameter.

We want: average of all possible sample variances

= population variance

• This is called unbiased .

Bessel’s game

1.

2.

List all possible samples of 2 cards.

Calculate sample averages.

Sample

Sample average

Population of all cards in a bag

Bessel’s game

1.

2.

List all possible samples of 2 cards.

Calculate sample averages.

3.

4.

Now, half of you calculate sample variance using /n, and half of you using /(n-1).

And then average all sample variances.

Population of all cards in a bag 𝑠 2

π‘₯ 𝑖

− π‘₯ 2

= 𝑛 OR 𝑛 − 1

Sample

0,2

Sample average

Sample variance

1

0,4 2

2,0

2,4

4,0

4,2

0,0

2,2

4,4

1

3

2

3

0

2

4

Sample

0,2

0,4

2,0

2,4

4,0

4,2

0,0

2,2

4,4 average

Bessel’s game πœ‡ = 2, 𝜎 2 =

8

3

Sample average

1

2

1

3

0

3

2

2

4

18

= 2

9

Sample variance (n-1) Sample variance (n)

2

0

2

8

0

2

8

2

0

24

=

9

8

3

1

4

1

1

4

1

0

0

0

12

=

9

4

3

Median absolute deviation (MAD)

• standard deviation is not robust

• IQR is robust

• mean absolute deviation MAD – a robust equivalent of the standard deviation

• Také your data, find median, calculate absolute deviation from the median, find the median of absolutes deviations

Median absolute deviation (MAD)

5

15

10

15

Data Median deviation Absolute deviation

5

10

30

20

30

Median:

MAD:

NORMAL

DISTRIBUTION

Playing chess

• Pretend I am a chess player.

• Which of the following tells you most about how good I am:

1.

2.

3.

My rating is 1800.

8110 th place among world competitive chess players.

Ranked higher than 88% of competitive chess players.

Distribution

We should use relative frequencies and convert all absolute frequencies to proportions.

Distribution of scores in one particular year

Height data – absolute frequencies http://wiki.stat.ucla.edu/socr/index.php/SOCR_Data_Dinov_020108_HeightsWeights

Height data – relative frequencies

Height data – relative frequencies

What proportion of values is between 170 cm and 173.75 cm?

30%

Height data – relative frequencies

What proportion of values is between

170 cm and 175 cm?

We can’t tell for certain.

• How should we modify data/histogram to allow us a more detail?

1.

2.

3.

Adding more value to the dataset

Increasing the bin size

A smaller bin size

Height data – relative frequencies

What proportion of values is between 170 cm and 175 cm?

36%

Height data – relative frequencies

Height data – relative frequencies

Normal distribution recall the empirical rule

68-95-99.7

1

2πœŽπœ‹ 𝑒π‘₯𝑝 − π‘₯ − πœ‡

2𝜎 2

2

Download