POL242Y5 Lecture Notes - Lecture 15: Uncorrelated Random Variables, Regression Analysis
Document Summary
For the independent variable, it can be any kind of variable but for the dependent it should be an interval. For formulating a hypothesis, keep things very simple. Different ideas require different methods and a regression analysis cannot test all of the ideas we have. So, when conducting a hypothesis consider applying certain methods when developing your ideas. Additionally, make sure the hypothesis is between two variables and nothing more than that. It can go from -1 to 1 and this will determine how strongly these variables are correlated/associated. If it is positive, it means the two variables are positively correlated. For regression, it tells us, on average, how much a one unit increase in the independent variable increases/decreases the value of the dependent variable. This is why regression provides us with a more precise understanding of the strength of the relationship.