So I’m taking the math level 2 sat subject test this October and decided to study topics on the Test one by one and have fewer problems on the exam day. I was wondering through the topics and I found this ( least squares
regression (linear, quadratic,exponential) ) Under the Data analysis, statistics, and probability. The thing is that I never learnt this topic in school neither can I find it on khan academy and such websites. So can anyone tell me what’s that and where can I learn more about it and the way It’s gonna be asked in the test ?
Regression is used to model a relationship between two sets of data. This image perfectly summarizes Regression:
Sets of data can show trends, and these trends can be modeled with functions. For example, a graph that takes the shape Y = aX + B, then there is Linear Regression. When I took he Math Level 2 SAT, I didn’t see a single question on Regression, and the most they would make you do is recognize what type of regression represents some graph or vice versa. You might just need to know this:
Linear Regression: y = a0 + a1x where a0 is the y-intercept and a1 is the rate of change
Quadratic Regression: y = a0 + a1x + a2x^2
Exponential Regression: y = a0e^(a1x)
Basically, regression is used to estimate data not given. For example, if there is a graph of years vs sales and they give you the sales for 2000, 2005, 2010, and 2015 but want you to find the sales of 2012, then you would find the regression function then plug in “2012”. However, this requires you to to use graphing calculator and input all your data and is quite time consuming, so I doubt you would ever see a question like that. If you want to read more info, Barron’s SAT Math Level 2 has a page or two with how to tackle such a complex problem. Good luck!
Isn’t that scatter plots? I mean i know all about scatter plots so if that’s the case i can tackle it :3
@Nabeel12 yes, pretty much. A linear regression is simply a line that best represents the data. Often we use the least squares regression, which is the line that minimizes the sum of the squares of the “residuals” (difference between observed value and value predicted by the regression line).
Most graphing calculators can perform regressions. Math II doesn’t go too deep into linear regressions (you likely won’t be asked to compute correlation coefficients, etc.).