28 Flashcards
Correlation linear equation?
y = beta_0 + beta_1(x)
Correlation quadratic equation?
y = x^2
Formula of linear correlation coefficient of random variables X and Y is?
rho_X,Y = (cov(X, Y)/(sigma_X * sigma_Y)), sigma_X * sigma_Y is standard deviation for each variable
Linear correlation coefficient of random variables X and Y?
This can determine strength and direction of linear association between two variables.
Covariance measures what in correlation?
How the variables vary together. Measures the tendency of two numerical variables to change together along a straight line.
If random variables X and Y are independent, when how does this affect the correlations of Covariance and Rho?
They will both be equal to 0 with no correlation.
Covariance correlation for point estimates?
cov(x,y)
Linear correlation coefficient for point estimator?
r = cov(x, y)/(s_x * s_y)
Why can’t I use covariance to talk about strength?
Covariance does not have bounds.
In simple linear regression, what is examined?
Mean response (mu_Y|x) variable Y and predictor variable X.
Mean response (mu_Y|x)?
What do you expect on average for your response when you have a particular X as your predictor
!!!Predictor?
Assumptions in simple linear regression?
•epsilon_i are independent normal random variables with a mean of zero and common variance sigma^2
•beta_0 and beta_1 are parameters
•X_i are not a random variable.
Error in simple linear regression, epsilon, has what kind of distribution?
Normal distribution where the mean touches the line of best fit throughout the entire line where only the mean changes but not the variance and standard deviation.