# for this class, you will be using the iris and mtcars datasets



# TASK 1
# descriptive statistics --------------------------------------------------
# Use the summary(), mean(), sd(),... function to summarize the iris dataset. Subset the dataframe for the different species
# such that you obtain, for example, mean +/- standard deviation of Sepal Length for the three different iris species.
# Next, plot a histogram to inspect the distribution of, for example, Petal Width for the three different species. 
# Explore the libraries pastecs (pastecs:stat.desc) and psych (psych:describeBy) for a number of descriptive statistics. 



# TASK 2
# correlation -------------------------------------------------------------
# First plot Sepal Length against Sepal Width for iris setosa.
# Next, plot histograms and use Shapiro's Test for normality to check for normal 
# distribution of Sepal Length and Sepal Width of iris setosa. Normal distribution of data is an important assumption.
# Use cor and cor.test to explore the correlation between sepal length and width for I. setosa.
# explore psych:pairs.panels



# TASK 3
# tests for differences in means ------------------------------------------
# Generally, t-tests are used to compare differences between two groups. ANOVA is used to compare differences among more than two groups.
# Use t-tests to test for statistical significance of differences in sepal length between Iris setosa and Iris versicolor.
# First use boxplot() for visualization of sepal length of I. setosa and I. versicolor.
# Check for normal distribution
# An important assumption for t-tests is the homogeneity of variance of the two datasets. Use var.test() to check for homoscedasticity.
# Finally, use t.test() to check for statistical significance of the difference in  in sepal length between Iris setosa and Iris versicolor.
# Use ANOVA to test for differences in sepal length of the three iris species. Check online for ANOVA in R. Hint: aov
# USe boxplot to visualize sepal length of the three iris species.



# TASK 4
# linear models -----------------------------------------------------------
# We use the mtcars dataset. Use descriptive statistics and associated visualizations to explore this dataset.
# Linear regressions are a special form of linear models. In R, linear models are formalized like this: lm(x~Y), which could be tranlated to: "X depends on Y".
# Follow the    mod<-lm(X~Y)  to build a first linear model between mpg (miles per gallon) and hp (horsepower)
# Use summary(mod) to inspect model outputs. Check online for interpretation. 
# Check online for visualization of the model result. Hint: abline()