This is 1/2 of my EXTENSIVE notes for Test 1 for the pre-master course Inferential Statistics on the University of Twente. With these notes, and the second part, you have a good knowledge to pass the first exam. I passed my exam with a 8.2! These notes consists of lecture 1 - 3. For lectures 4 - 6,...
Inhoudsopgave
Notes Inferential Statistics – Test 1 ....................................................................................................................... 2
Self-Study B: the normal model + finding probabilities ...................................................................................... 2
Lecture 1: Introduction to statistical inference. Sampling Distribution Model for A PROPORTION and
Confidence intervals for PROPORTIONS........................................................................................................... 8
Statistical inference and the fundamental problem of uncertainty in statistics ............................................... 8
Sampling distribution model FOR PROPORTIONS (theory + assumptions) ................................................... 10
Using the theory about the Sampling Distribution Model for PROPORTIONS (theoretical examples, canvas
page 1.1: ex1 (first year student), ex3 + ex4) .................................................................................................... 13
5. Using the theory about the Sampling Distribution Model for PROPORTIONS in practice: → Confidence
interval for a proportion .................................................................................................................................... 15
1.1: Sampling distribution model for proportions ............................................................................................. 17
Exercises 1.1 Canvas ......................................................................................................................................... 17
Lecture 2: Inference about one mean: Confidence intervals (SDM + CLT + TEST) .................................... 23
Sampling distribution model for a mean............................................................................................................ 23
Inference for means: confidence interval for a mean ........................................................................................ 26
2.1: Sampling distribution model for means (SDM + CLT + TEST) .............................................................. 29
Exercises Canvas ............................................................................................................................................... 29
2.2: Confidence interval for a mean .................................................................................................................... 33
Exercises Canvas ............................................................................................................................................... 35
Reflection Session 1: CI Proportions and Means .............................................................................................. 39
Lecture 3: Inference for means – Test for one mean......................................................................................... 45
Inference for means: hypothesis test for one mean............................................................................................ 45
An example of the one sample t-test................................................................................................................... 48
So an conclusion / overview t-procedure for the one sample situation: ............................................................ 55
Lecture 3B: More about tests .............................................................................................................................. 56
Interpretation of outcomes (statistical significance, magnitude (grootte) of the outcome and practical
significance) ....................................................................................................................................................... 56
Investigating normality: Descriptive + Shapiro-Wilk test................................................................................. 57
More about assumptions and an alternative solution (when to use the Sign test)............................................. 60
Canvas Lecture 3B: more about tests – part II.................................................................................................. 61
1
,3.1: Steps in performing a statistical test ............................................................................................................ 68
3.3: Exercises More about Tests .......................................................................................................................... 82
Notes Inferential Statistics – Test 1
Self-Study B: the normal model + finding probabilities
Normal model = N(mu, sigma)
Standard normal model (or: Z-distribution) = N(0,1)
Exercises via the distribution of IQ-scores
Assume: normal distribution
- Mean = 100
- SD = 16
→ IQ has a N(100, 16)-distribution
1. What percentage of observations is expected to fall within one standard deviation
from the mean and what are the borders of that interval?
→ 68% of values fall within 1 SD from the mean
→ 68% of the values fall between: 100 – 16 = 84 & 100 + 16 = 116
2. What percentage of observations is expected to fall within two standard deviations
from the mean and what are the borders of that interval?
→ 95% of values fall within 2 SD from the mean
→ 95% of the values fall between: 100 – 32 = 68 & 100 + 32 = 132
3. What is the expected proportion that has at least a score of 132 or more on the IQ-
scale?
→ IQ-score of 132 is at least 2 SD away from the mean.
→This means that 100 – 95 = 5% / 2 = 2,5% has at least a score of 132 or more.
Using the N(0, 1)-distribution and reading the z-table
Calculating z-scores by:
(observation – mean of population (mu) ) / SD (sigma)
Proportions on Z-table = part below
1 - proportions on Z-table = part above
4. How big is the group with an IQ-score of 124 or more? (What is the expected
proportion that has at least a score of 124 or more on the IQ-scale?)
▪ 4a. What is the standardized score, the z-value, for the IQ-score 124?
→ 124 – 100 (mean) / 16 (SD) = 1.5
▪ 4b. What is the expected proportion that has at least a score of 124 or more on the IQ-
scale? (assuming a normal model)
2
, (use Z-table with Standardized Normal Probabilities)
→ So P(Z > 1.5) = ?
→ In the Z-table you see when Z is 1.5, the SNP is .9332.
→ Expected proportion that has 124+ IQ = 1 – 0.9332 = 0.0668 = 6.68%
Other examples
- What percent of a standard normal model is found for: z < 2.25?
→ .9878 (see Z-table) = 98.78%
- What percent of a standard normal model is found for: -1 < z < 1.15?
→ 1.15 = .8749 & -1 = .1587
→ .8749 - .1587 = .7162 = 71.62%
- Reversing the problem. How can we find the z-value for a certain cumulative
probability?
o What is the value of Z for the highest 75%? →
100 – 75 = 25% = 0,2500.
Z-table says .2514 is closest, so Z-score = -.67
Z-table says there are two values near the probability of .9500, namely .9495 with a Z-
value of 1.64 & .9505 with a Z-value of 1.65. --> We take the z-value in between:
1.645
Specific value for the length
→ z = (y – mu) / sigma
→ 1.645 = (y – 1.76) / .063 = (1.645 * .063) + 1.76 = 1.864 meter
Exercise 1: Temperatures
The high temperature in a town seems to be 2 degrees Celsius on average in January, with a
standard deviation of 6 degrees. In July the mean high temperature is on average 24 degrees
Celsius, with a standard deviation of 5 degrees. In which month is it more unusual to have a
day with a high temperature of 13 degrees?
→ Distance is for both the same, but the probability (normal model) is not the same
o 13 – 2 (= mean January) = 11 & 24 (= mean July) – 13 = 11
→ In January, with mean 2 and SD 6, a high temperature of 13 is almost 2 standard deviations
above the mean ( = 1.83 SD)
→ in July, with mean 24 and SD 5, a high temperature of 13 is more than two standard
deviations below the mean ( = 2.2 SD)
→ So a high temperature of 13 degrees Celsius is less likely to happen in July, when 13
degrees Celsius is farther away from the mean
Exercise 1b: Temperatures
A town’s January high temperatures average 36F with a standard deviation of 10, while in
July the mean high temperature 74 and the standard deviation is 8. In which month is it more
unusual to have a day with a high temperature of 55? Explain.
→ Distance is for both the same, but the probability (normal model) is not the same
o January = 55 – 36 = 19 & distance July = 74 – 55 = 19
3
, → In January, with mean 36 and SD 10, a high temperature of 55 is almost 2 standard
deviations above the mean (19/10 = 1.9 SD)
→ In July, with mean 74 and SD 8, a high temperature of 55 is more than two standard
deviations below the mean (19/8 = 2.375 SD)
→ so a high temperature of 55 is less likely to happen in July, when 55 is farther away from
the mean
Exercise 2: Length of students
- Mean = 1.76 meter
- SD = 0.63 meter
- Assumed we can apply normal model
a. Draw a model for length, label it in terms of mu and sigma, and make clear what the
68-95-99.7 rule predicts
Standard numbers!:
- Between plus 1 and minus 1 = 68%
- Between 0 and plus OR minus 1 = 34%
- Between plus 1 and plus 2 OR minus 1 and minus 2 = 13.5%
- Between plus 2 and plus 3 OR minus 2 and minus 3 = 2.5%
- Between plus 1 and plus 3 OR minus 1 and minus 3 = 16%
- Between 0 and plus 3 OR 0 and minus 3 = 50% (49.85)
- Plus 3 standard deviations away from the mean = 1.9949 meters
- Minus 3 standard deviations away from the mean = 1.571 meters
b. Give the interval you would expect for the central 68% in terms of length
→ between 1.697 and 1.823 meter
c. About what percentage of students do you expect to have a length of more than 1.823
meter?
→ 100 – 68 = = 16%
d. About what percentage of students do you expect to have a length between 1.823 and
1.886 meter?
→ 16% - 2.5% = 13.5%
e. What is the length of the shortest 2.5% of students?
→ 1.634 meters or shorter
4
The benefits of buying summaries with Stuvia:
Guaranteed quality through customer reviews
Stuvia customers have reviewed more than 700,000 summaries. This how you know that you are buying the best documents.
Quick and easy check-out
You can quickly pay through credit card or Stuvia-credit for the summaries. There is no membership needed.
Focus on what matters
Your fellow students write the study notes themselves, which is why the documents are always reliable and up-to-date. This ensures you quickly get to the core!
Frequently asked questions
What do I get when I buy this document?
You get a PDF, available immediately after your purchase. The purchased document is accessible anytime, anywhere and indefinitely through your profile.
Satisfaction guarantee: how does it work?
Our satisfaction guarantee ensures that you always find a study document that suits you well. You fill out a form, and our customer service team takes care of the rest.
Who am I buying these notes from?
Stuvia is a marketplace, so you are not buying this document from us, but from seller Jhamel. Stuvia facilitates payment to the seller.
Will I be stuck with a subscription?
No, you only buy these notes for $9.77. You're not tied to anything after your purchase.
