ISYE 6501X EXAM PREP TEST BANK
VERIFIED QUESTIONS AND CORRECT
ANSWERS
●● Box-Cox Transformation
Answer: A logarithmic transformation used to transform non-normal
response data into a form that is closer to normally distributed.
●● Eigenvalues
Answer: In Principal Component Analysis (PCA), these values are used
to sort eigenvectors; they represent the amount of variance captured by
each principal component.
●● Eigenvectors
Answer: Vectors calculated during PCA that define the new coordinate
system; the data is transformed into these new factors to remove
correlation.
●● Heteroscedasticity
Answer: A condition where the variance of data points is unequal across
the range of values (e.g., higher variance at higher values), which can
bias regression models.
, ●● K-Nearest Neighbor (k-NN) Regression
Answer: A non-parametric method that predicts a response for a new
data point by averaging the responses of the k closest data points in the
existing dataset.
●● Knot
Answer: The connection point in a regression spline where different
polynomial functions meet, requiring smooth transitions between the
pieces.
●● Orthogonal
Answer: A term describing the new coordinate system in PCA where the
directions are at right angles to each other, resulting in zero correlation
between them.
●● Poisson Regression
Answer: A regression method used when the response follows a Poisson
distribution (e.g., counting arrivals), where the parameter lambda is
modeled as a function of attributes.
●● Principal Component Analysis (PCA)
Answer: A feature extraction method that transforms high-dimensional,
correlated data into a ranked set of uncorrelated (orthogonal)
components.
VERIFIED QUESTIONS AND CORRECT
ANSWERS
●● Box-Cox Transformation
Answer: A logarithmic transformation used to transform non-normal
response data into a form that is closer to normally distributed.
●● Eigenvalues
Answer: In Principal Component Analysis (PCA), these values are used
to sort eigenvectors; they represent the amount of variance captured by
each principal component.
●● Eigenvectors
Answer: Vectors calculated during PCA that define the new coordinate
system; the data is transformed into these new factors to remove
correlation.
●● Heteroscedasticity
Answer: A condition where the variance of data points is unequal across
the range of values (e.g., higher variance at higher values), which can
bias regression models.
, ●● K-Nearest Neighbor (k-NN) Regression
Answer: A non-parametric method that predicts a response for a new
data point by averaging the responses of the k closest data points in the
existing dataset.
●● Knot
Answer: The connection point in a regression spline where different
polynomial functions meet, requiring smooth transitions between the
pieces.
●● Orthogonal
Answer: A term describing the new coordinate system in PCA where the
directions are at right angles to each other, resulting in zero correlation
between them.
●● Poisson Regression
Answer: A regression method used when the response follows a Poisson
distribution (e.g., counting arrivals), where the parameter lambda is
modeled as a function of attributes.
●● Principal Component Analysis (PCA)
Answer: A feature extraction method that transforms high-dimensional,
correlated data into a ranked set of uncorrelated (orthogonal)
components.
