4 posts with this tag
In the last post I talked about how to find the coefficients that give us the line of best fit for a OLS regression problem using the normal solution. The core of this approach is the equation: $$ X^TXb = X^Ty $$The way we solved this in the previous …
When I first learned least-squares linear regression in my undergrad degree, I remember that we approached it in the "calculus" way: taking the sum of the squared differences for each observation and solving a massive (and tedious) equation until we …
Choosing the right linear regression model for your data can be an overwhelming venture, especially when you have a large number of available predictors. Luckily R has a wide array of in-built and user-written tools to make this process easier. In …
Coming from a psychology background, I have a soft spot for multiple linear regression. This method is a workhorse in statistics and machine learning, being flexible, powerful and easily interpretable. An issue that people new to multiple regression …
