MISY261 Midterm: Exam 3 Prep - Videos

5 Pages
Unlock Document

Management Information Systems
Alex Chaplin

Exam 3 Prep Videos Logistic Regression: Introduction Linear regression predicts continuous outcomes Binary logistic regression helps with yesno outcomes Logistic Regression: Setup Probability of success = p Probability of failure = 1p Defined the problem with just one unknown value, p One of the possible outcomes is always assigned as success value Construct a linear model of the odds ratio log( ) = + + + + 1 0 1 1 o Right hand side is identical to linear regression model o ( ) 1 o R assumes that the log is a log with a base e o This function will help us classify 0s and 1s more easily Odds Ratio Odds ratio is Odds ratio is not the same as the probability o success o Probability of success = p, odds ratio =1 o When the probability of success increases, the odds ratio increases Odds ratio can be >1 o If this is the case, it means the probability of success is greater than the probability of failure If the odds ratio = 1, this means the probability of success = the probability of failure Logarithm Review Log = function that takes a number and returns a number Cannot take the log of a number 0 o Odds ratio will never be negative because it is made of probabilities Log(1)=0, so if there is an odds ratio of 1, the result will be 0 o If the odds ratio is <1, then the log(odds ratio) will be <0 o If the odds ratio is >1, the log(odds ratio) will be >0 Logistic Regression in R (Implementation) In R, change lm to glm o Generalized linear model o Fewer assumptions o Takes more arguments o R uses Maximum Likelihood Estimation to try out a bunch of numbers to see what are the most likely slopes and coefficients Finds the best model o Ex: fit = glm(doptionsexp ~ dMarketVolume + dVIXclose, family=binomial doptionsex binary response (the thing youre trying to predict)
More Less

Related notes for MISY261

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.