lecture 4 confidence intervals. lecture summary last lecture, we talked about summary statistics and...
DESCRIPTION
IntroductionTRANSCRIPT
![Page 1: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/1.jpg)
Lecture 4
Confidence Intervals
![Page 2: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/2.jpg)
Lecture Summary
• Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters– Risk, bias, and variance– Sampling distribution
• Another quantitative measure of how “good” the statistic is called confidence intervals (CI)
• CIs provide an interval of certainty about the parameter
![Page 3: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/3.jpg)
Introduction
• Up to now, we obtained point estimates for parameters from the sample – Examples: sample mean, sample variance, sample median,
sample quantile, IQR, etc.– They are called point estimates because they provide one
single point/value/estimate about the parameter– Mathematically: a single point!
• However, suppose we want a range of possible estimates for the parameter, an interval estimate like [– Mathematically: and
![Page 4: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/4.jpg)
Two-Sided Confidence Intervals• Data/Sample: – is the parameter
• Two-Sided Confidence Intervals: A -confidence interval is a random interval, from the sample where the following holds
– Interpretation: the probability of the interval covering the parameter must exceed
– It is NOT the probability of the parameter being inside the interval!!! Why?
Confidence Level
![Page 5: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/5.jpg)
![Page 6: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/6.jpg)
Comments about CIs• Pop quiz 1: What is the confidence level, for confidence
interval?– Thus, for any level CI would be a valid (but terrible) CI
• Pop quiz 2: What is the confidence level, , for CI where is any number?
• Pop quiz 3: Suppose you have two confidence intervals and . If the first CI is shorter than the second, what does this imply?– If is the same for both intervals, what would this imply about the
short interval (in comparison to the longer interval)?
• Main point: given some confidence level , you want to obtain the shortest CI
![Page 7: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/7.jpg)
CIs for Population Mean
• Case 1: If the population is Normal and is known
CI: Hint: Use sampling distribution of
• Case 2: If the population is not Normal and is known
Approximate CI: Hint: Use CLT of
![Page 8: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/8.jpg)
What if the variance,,is unknown?
![Page 9: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/9.jpg)
t Distribution• Formal Definition: A random variable has a t-distribution
with degrees of freedom, denoted as , with the probability density function
where is a gamma function
• Useful Definition: Consider the following random variable
where and and are independent. Then,
• “Quick and Dirty” Definition: If , i.i.d., then
You will prove the relation between the two in the homework
Notice that you can transform into a standard Normal
From lecture 3, is with some constant multipliers
![Page 10: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/10.jpg)
![Page 11: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/11.jpg)
Property of the t Distribution• The t-distribution has a fatter tail than the normal distribution (see
picture from previous slide)– Consequences: The “tail” probabilities for the t distribution is bigger than
that from the normal distribution!
• If the degrees of freedom goes to , then
Proof: CLT!
• This means that with large sample size , we can approximate with a standard normal distribution
for large – General rule of thumb for how large should be:
![Page 12: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/12.jpg)
CIs for Population Mean
• Case 3: If the population is Normal and variance is unknown
CI: Hint: Use the “quick and dirty” version of the t-
distribution• Case 4: If the population is not Normal and variance is
unknown (i.e. the “realistic” scenario)Approximate CI: Hint: Use CLT! – Demo in class
![Page 13: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/13.jpg)
![Page 14: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/14.jpg)
Summary of CIs for the Population MeanScenarios CI Derivation
1) Population is Normal2) Variance is known
Use sampling distribution for
1) Population is not Normal2) Variance is known
Approximate CI, use CLT
1) Population is Normal2) Variance is unknown
Use the t distribution
1) Population is not Normal2) Variance is unknown
Approximate CI, use CLT
Fixed width CI
Variable width CI
![Page 15: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/15.jpg)
CIs for Population Variance
• Case I: If the population is Normal and all parameters are unknown
[ ]– Hint: Use the sampling distribution for
• Case II: (Homework question) If the population is Normal and the population mean is known.– Hint: Use and the sampling distribution related to it!
![Page 16: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/16.jpg)
Lecture Summary
• Another quantitative measure of how “good” the statistic is called confidence intervals (CI)
• CIs provide an interval of certainty about the parameter
• We derived results for the population mean and the population variance, under various assumptions about the population – Normal vs. not Normal– known variance vs. unknown variance
![Page 17: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/17.jpg)
Extra Slides
![Page 18: Lecture 4 Confidence Intervals. Lecture Summary Last lecture, we talked about summary statistics and how “good” they were in estimating the parameters](https://reader030.vdocument.in/reader030/viewer/2022012914/5a4d1b577f8b9ab0599a9d32/html5/thumbnails/18.jpg)
One-Sided Confidence Intervals
• One-Sided Confidence Intervals: A -confidence interval is a random interval, from the sample where the following holds
• One-Sided Confidence Intervals: A -confidence interval is a random interval, from the sample where the following holds