Reading 11: Sampling and Estimation

Question 1

What is simple random sampling, sampling distribution, and a sampling error?

Answer 1

Simple random sampling- is the art of choosing samples from a population at random to represent the population.

Sampling Distribution- is the probability distribution function that takes those samples and makes it to a sampling curve.

Sampling Error of the Mean= Sample Mean- Population Mean= x- u

Question 2

Tell me the difference between sampling variations. I want to know the difference between stratified random sampling and simple random sampling?

Answer 2

Stratified Random sampling is when you take the population and divide the population into segments based on 1 or more distinguishing characters. Choose a select number from each strata.

Simple Random sampling- you choose at random a bunch of simple random samples.

Question 3

What is the difference between these 2 data collection methods: Time-Series and Cross-Sectional Data? Longitudinal Data?

Answer 3

Time Series Data- time series data is a set of monthly returns on Microsoft for years 1999-2004

Cross-Sectional Data- is data collected at a single point in time for a variety of different subjects.

EX: DEC 2004 EPS for all NASDAQ Market Stocks

Longitudinal Data- is multiple recordings of DIFFERENT CHARACTERISTICS of a single identity over a period of time.

EX of Longitudinal Data:
EPS, Market Share, P/B Ratio of NOKIA over 10 years

Question 4

Please explain the Central Limit Theorem and why its important. OK?

Answer 4

Central Limit Theorem is important because:

CLT is used to check out the influence of sample size on the SAMPLING ERROR. The sampling error is actually getting smaller as the sample size increases.

***What is Sampling Error? Sampling Error is the difference between:

sample mean and population mean

Question 5

How can we calculate the sampling error?

Answer 5

SD ERROR sample= SD pop/ (n)^.5

Question 6

Please describe the characteristics that make a population estimator desirable?

Answer 6

Characteristics include:

unbiasedness- how close is the sample mean to pop mean?

efficiency- how small is the variance?

consistency- accuracy will always increase as sample size increases.

Question 7

Let’s understand what is a T-distribution and its properties?

Answer 7

T-distributions are characterized as:

1. ) having less than 30 samples
2. ) unknown population variances
3. ) data-set must be normally distributed

DF= N-1 (As DF increases, it approaches a normally distributed curve)

Question 8

How do you calculate Confidence Intervals? What is a Confidence Interval?

Answer 8

Confidence Interval= is the confidence that given a set CONFIDENCE LEVEL OF 95%,99%,ETC.

**CI= Xbar + (Za/2) (SD/ square root of n)