Excel Statistical Analysis 41: Confidence Interval for t Distribution, use when Sigma NOT Known

preview_player
Показать описание
Learn about how to create Confidence Interval to estimate a population Mean when Sigma (Population Standard Deviation) is NOT Known using the t Distribution and the Excel worksheet functions: T.INV, CONFIDENCE.T and also the Data Analysis tool.
Topics:
1. (00:00) Introduction
2. (00:27) t Distributions: Characteristics
3. (04:20) t Distribution Confidence Interval Formula
4. (04:48) Excel T Functions
5. (05:26) How t Curve changes as the Sample Size, n, changes
6. (06:34) Printer Cartridge Example With formulas to calculate sample statistics
7. (08:55) Method #1: T.INV function
8. (09:54) Method #2: CONFIDENCE.T function
9. (11:15) Method #3: Data Analysis Descriptive Statistics Feature
10. (14:27) Restaurant Rating Example, All Three Methods Demonstrated
11. (18:10) Summary of video
12. (18:28) Closing, Next Video and Video Links
Рекомендации по теме
Комментарии
Автор

Well oranized and carefully explained! There are obviously 2 issues addressed by normalized t relative to normaized z: unknown SD and small sample size. However the CLT tends to make BOTH less of an issue by increasing sample size. If you plot the t.dist norm.dist as anoverlay and the ibcrement sample sizebit is amazing how fast they converge as "n". Increases. By n=30 orvl so the convergence is almost complete and even at n=20 It's easy to say "why bother"!! But probably you should until at least n=30. Thus it really comes down to the single issue of small samples. But the most frequently asked question by my beginning students was why STUDENT'S t? And they found it even mote confusing when I tlold them it was all about of BEER!! At the end of the 19th and the early 20th century the Big Guy of statistics was Karl Pearson whose name still populates most Psychology texts. Pearson was smart bu he was confused about how to deal with small samples l. Large samples were expensive and hard to process especially then. There was an Oxford educated guy named Bill Gosset who was hired by the Guinness company to give it a competitive edge in quality control, etc. Bill G. would finish his day and read about what smart stuff Kark P. was up to and decided he could solve Pearson's problem of how small samples relate to the CLT. He wrote an article worthy of publication BUT the Guinnees brass freaked out since the fact that a statistician was helping Guinness beat its competition was top secret. They let him publish but not under his real name. Bill G. became "Student-t" so that he could contribute to statistics far more than he ever did to Beer!

richardhay
Автор

Boom!Another Super Cool Class...Thank You Mike :)

darrylmorgan
Автор

Hi Mike, Thanks for making video for us. For some issue YouTube didn't allowed to post my comments on your previous videos had to flash my phone

msantosh
Автор

Hi Mike, what is the name of the textbook you are using for the class?

darylseaton