Contents
Week 1 .................................................................................................................... 2
Week 2 .................................................................................................................... 5
Week 3 .................................................................................................................. 11
Week 4 .................................................................................................................. 18
Week 5 .................................................................................................................. 24
Week 6 .................................................................................................................. 28
Week 7 .................................................................................................................. 34
,Week 1
Part 1: Basics
Statistical reasoning:
You cannot only recommend something only with the mean. You need to consider
variability. With a low standard deviation is better
Statistical modelling:
The larger the test statistic the better!
Sampling distribution = the distribution of the test statistic (histogram)
Distribution of a random variable = the possible values of a variable (for example:
age,gender,income,etc.)
P value = the probability of observing a test statistic at least as large as estimated t, if H0
is true.
,Conclusion
Part 2: Design
Steps:
1. Set a research question
2. Convert the question into some type of model or test
a. Is the problem supervised or unsupervised?
b. Is the problem inference-related or prediction-related?
, c. Are there any limitations?
3. Collect the necessary data
a. What variables are necessary?
b. Are any key constructs very rare?
c. Do you need experimental data?
d. What sample size is required?
4. Exploratory Data Analysis (EDA) = a way to interactive analyse/explore data
Week 1 .................................................................................................................... 2
Week 2 .................................................................................................................... 5
Week 3 .................................................................................................................. 11
Week 4 .................................................................................................................. 18
Week 5 .................................................................................................................. 24
Week 6 .................................................................................................................. 28
Week 7 .................................................................................................................. 34
,Week 1
Part 1: Basics
Statistical reasoning:
You cannot only recommend something only with the mean. You need to consider
variability. With a low standard deviation is better
Statistical modelling:
The larger the test statistic the better!
Sampling distribution = the distribution of the test statistic (histogram)
Distribution of a random variable = the possible values of a variable (for example:
age,gender,income,etc.)
P value = the probability of observing a test statistic at least as large as estimated t, if H0
is true.
,Conclusion
Part 2: Design
Steps:
1. Set a research question
2. Convert the question into some type of model or test
a. Is the problem supervised or unsupervised?
b. Is the problem inference-related or prediction-related?
, c. Are there any limitations?
3. Collect the necessary data
a. What variables are necessary?
b. Are any key constructs very rare?
c. Do you need experimental data?
d. What sample size is required?
4. Exploratory Data Analysis (EDA) = a way to interactive analyse/explore data
