BADM 211 Exam 2 – Questions With Final
Answers
Save
Terms in this set (91)
what doesnt effectively Mean Error: absolute error better at displaying the
asses predictive true range if negative values and positives average
performance? to 0
-Linear functional form
-Flexible; can be applied for predictive and
MLR:
explanatory modeling
-Caution: we need to avoid collinearity
if the predictor p_i, i <.05: Yes
would it be considered
significantly associated
with the outcome
variable?
if beta_i > 0 then what postive
kind of association is
there between the
predictor variable and
the outcome variable?
- that there is a linear relationship between the
predictors and the outcome variable
- the cases are independent of each other
Assumptions of MLR
- the noise follows a normal distribution
-the variance does not depend on the values of the
predictors
-the aim of partitioning data is to avoid this
-occurs when we overestimate the model
Overfitting
performance based on the data to the training set
-partition the data into training and validation sets
, 1- subtracting the mean and dividing by the
Two ways of normalizing
standard deviation
the data are
2- rescale the variation to uniform range to 0to1
In a dataset, rows and rows = y axis of data points/observations
columns correspond to columns = x axis of variables/catrgories
________ and _________
respectively.
A numerical variable can where the measurement or number has a numerical
be defined as ___________. meaning
For nominal data! Set list of variable options where
We need to dummy code the final option would be implied
EX: gender
Two types of categorical nominal - not ordered
variables are ____________ ordinal - ordered and ranking
and
how much one variable is affected by another. The
Scatter plots represent relationship between two variables is called their
correlation
We use _________ to boxplots and histograms
visualize the entire
distribution of a variable
Linear regression is a statistical technique where the
score of a variable Y is predicted from the score of a
What is linear regression?
second variable X. X is referred to as the predictor
variable and Y as the criterion variable.
What are the differences explanatory is evaluating how well the predictive
between predictive model explains an outcome
modeling and
explanatory modeling?
Answers
Save
Terms in this set (91)
what doesnt effectively Mean Error: absolute error better at displaying the
asses predictive true range if negative values and positives average
performance? to 0
-Linear functional form
-Flexible; can be applied for predictive and
MLR:
explanatory modeling
-Caution: we need to avoid collinearity
if the predictor p_i, i <.05: Yes
would it be considered
significantly associated
with the outcome
variable?
if beta_i > 0 then what postive
kind of association is
there between the
predictor variable and
the outcome variable?
- that there is a linear relationship between the
predictors and the outcome variable
- the cases are independent of each other
Assumptions of MLR
- the noise follows a normal distribution
-the variance does not depend on the values of the
predictors
-the aim of partitioning data is to avoid this
-occurs when we overestimate the model
Overfitting
performance based on the data to the training set
-partition the data into training and validation sets
, 1- subtracting the mean and dividing by the
Two ways of normalizing
standard deviation
the data are
2- rescale the variation to uniform range to 0to1
In a dataset, rows and rows = y axis of data points/observations
columns correspond to columns = x axis of variables/catrgories
________ and _________
respectively.
A numerical variable can where the measurement or number has a numerical
be defined as ___________. meaning
For nominal data! Set list of variable options where
We need to dummy code the final option would be implied
EX: gender
Two types of categorical nominal - not ordered
variables are ____________ ordinal - ordered and ranking
and
how much one variable is affected by another. The
Scatter plots represent relationship between two variables is called their
correlation
We use _________ to boxplots and histograms
visualize the entire
distribution of a variable
Linear regression is a statistical technique where the
score of a variable Y is predicted from the score of a
What is linear regression?
second variable X. X is referred to as the predictor
variable and Y as the criterion variable.
What are the differences explanatory is evaluating how well the predictive
between predictive model explains an outcome
modeling and
explanatory modeling?
