Describe the following distributions:
-heterogeneity
-zero-inflated
-hurdle
-latent class - correct answer ✔✔A heterogeneity model requires modeling with a continuous mixture.
A zero-inflated model requires modeling as a mixture of a point mass at zero and another distribution
whose domain starts with 0.
A hurdle model requires modeling as a mixture of a point mass at zero and another distribution whose
domain starts with 1.
A latent class model uses a discrete mixture which models each subgroup with its own distribution.
What makes a model supervised? - correct answer ✔✔Has Response Variable
Two types of supervised learning and their basic quality - correct answer ✔✔Regression: Quantitative
response variable
Classification: Categorical response variable
Write out general formula for Test MSE - correct answer ✔✔see page 2
What three parts of linear models does error source from? - correct answer ✔✔Variance of model -
Reducible Error
Squared Bias - Reducible Error
Variance of residuals - Irreducible Error
What is the relationship between flexibility and interpretability? - correct answer ✔✔Inverse
, What happens to training MSE and test MSE as flexibility increases? - correct answer ✔✔training MSE
decreases, but test MSE follows a u-shaped curve
Low flexibility lead to.... And high flexibility leads to.... - correct answer ✔✔Low variance and high bias...
high variance and low bias
cov(x,y) is... - correct answer ✔✔see page 2
r(x,y) is - correct answer ✔✔see page 2
SLR- b1 = - correct answer ✔✔see page 3
SLR - b0 = - correct answer ✔✔see page 3
SLR - se,b0 = - correct answer ✔✔see page 3
SLR - se,b1 = - correct answer ✔✔see page 3
SLR - se,y-hat = - correct answer ✔✔see page 3
SLR - se, y-hat of n + 1 = - correct answer ✔✔see page 3
What matrix is X - correct answer ✔✔design
What matrix is H - correct answer ✔✔hat
What lies in the b matrix - correct answer ✔✔beta coefficients
H = - correct answer ✔✔X(X^(T)X)^(-1)X^(T)