Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Slopes of a Line, and Regression
#1
The only thing I remember from Algebra is that y=mx+b, and I don't remember what any of the letters stand for. Now, I'm minding my own business studying stats, and I see this formula in a book that looks something like this y(actually, it has a little carrot above it) = b0 (actually, the 0 is a subscript) + b1 (again, the 1 is a subscript)x. Now, they're calling it some sort of formula for linear regression, or something equally outlandish. Why are they doing this? Do the mathematics gods hate me? Not only do they expect me to learn a bunch of Greek letters (both upper and lower case), but they're trying to confuse me with different formulas that say the same thing. Or do they?

I guess that's my question. Are they the same formulas? And, if not, why not? If they are, you don't need to explain why two different forumlas are being used. I know. It's 'cause the math gods are trying to mess with my mind. banghead
TESU BSBA - GM, September 2015

"Never give up on a dream just because of the time it will take to accomplish it. The time will pass anyway." -- Earl Nightingale, radio personality and motivational speaker
Reply
#2
Stats can be confusing because you are going from certainty (y=b+mx) to, in a linear regression, a mathematical best guess model
(y-carrot=b(0) +b1x). Then on top of this you have an estimate, based on a normal distribution of error (i.e. the difference between your best guess and the actual value of y) so y(actual)=b(0) +b(1)x + e. Let's say you graph (called scatterplot) height as x and weight as y of a randomly (by chance) sample group of people, you would see on this graph that as height went up, weight went up (obviously expected of course), but not in an exact relationship (skinny tall people can be outweghed by short rotund people). Say you want to get a relationship of height vs weight (a carnival act maybe?) Here you're best guess (meaning the least range of error - that is how far off is your y(actual) - for all the y vs x in your sample data set - will be a linear regression model (a mathematical description for the phenomena of height vs weight)

Have you looked into the Prospero courses by Pearson:Propero They are self paced like SL but SL only has Business Stats instead of Math Stats which you need. Prospero will give you 24/7 online tutoring service by SmartThinking (TESC includes this service with any course you take from them). I believe you need a course for stats (You are expecting too much of yourself to do this on your own.) that is well organized (this is what guided study is all about vs. flashcard blitz) supplemented by a tutor that feels so good to answer your dumbest questions. A lot of confusion about stats is nomenclenture. Believe me, it's not consistent. So it's not you that is having the real problem, it's the teachers that can't get the issue simplified in a step by step manner. We'll just have to keep looking until we find what will work for you.

You have worked so hard and are so close. We are not going to abandon ship now!
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)