DATA MINING EXAM 1 CORRECT
100%
T/ F - Data Mining is a scientific approach to managerial decision making in which raw
date are processed and manipulated to produce meaningful information. -
ANSWERTrue
All data collected in a particular study is referred to what?
a) Elements
b) Data set
c) Variable
d) Observations - ANSWERData Set
T/F - Data are the facts and figures collected, analyzed, and summarized for
presentation and interpretation - ANSWERTrue
Which of the following is not part of the Scales of Measurement?
a. Nominal
b. Ratio
c. Interval
d. Data - ANSWERData
True/False: The first step in trying to reduce the number of predictors should always be
to use domain knowledge. - ANSWERTrue
True/False: Categorical variables can also be handled by most data mining routines and
do not require extra services. - ANSWERFalse
T/F: The less variables we include, the greater risk of overfitting data - ANSWERFalse
T/F: The goal of SUPERVISED (predictive) learning is to predict a single "target" or
"outcome" - ANSWERTrue
T/F: Regression analysis can be used to develop an equation showing how different
data variables are related. - ANSWERTrue
What are the 3 partitions typically used to address the overfitting problem when building
a predicting model? - ANSWERTraining, validation, and test
A set of measurements on an observation is
• A. Prediction
• B. Profile
• C. Variable
• D. Model - ANSWERProfile
, T/F Predicting the outcome value for new records, given their input values is a
predictive task. - ANSWERTrue
The goal of Supervised Predictive learning is to predict multiple "targets" or "outcome"
variables. - ANSWERFalse
What is the first step while testing for significance T-test?
1. select the test statistic
2. compute the value of the test statistic
3. determine the hypotheses
4. specify the level of significance - ANSWERdetermine the hypotheses
1) What does "e" stand for in the equation Y=B0 + B1*t + e
a) time b) error c) intercept d) trend - ANSWERError
When doing a weighted average the sum of weights must be greater than or equal to
one. - ANSWERFalse
In the Forecast Error Measurement MAD, n stands for a. period number b. actual
demand c. total number of periods d. forecast demand - ANSWERTotal number of
periods
T/F: A simple moving average formula implies equal weighting of prior time periods -
ANSWERTrue
1. (T/F) A short coming of a moving average is that it does not forecast seasonal
component. - ANSWERTrue
T/F: MSE is a measure of the average of the errors absolute value. - ANSWERFalse
T/F: Seasonality accounts for the gradual shifting of the time series over a long period of
time. - ANSWERfalse
T/F- To determine the seasonal factor of a time-series data set, it is important to
develop an average for the same quarters in each period and then find a seasonal
factor by dividing that average by the general average. - ANSWERTrue
T/F: the four components of time series is level, trend, seasonality, and noise. -
ANSWERTrue
The fifth step of decomposition is:
a) Deseasonalize the original data
b) Determine the seasonal factor
c) Create the final forecast by adjusting the regression line by the seasonal factor
100%
T/ F - Data Mining is a scientific approach to managerial decision making in which raw
date are processed and manipulated to produce meaningful information. -
ANSWERTrue
All data collected in a particular study is referred to what?
a) Elements
b) Data set
c) Variable
d) Observations - ANSWERData Set
T/F - Data are the facts and figures collected, analyzed, and summarized for
presentation and interpretation - ANSWERTrue
Which of the following is not part of the Scales of Measurement?
a. Nominal
b. Ratio
c. Interval
d. Data - ANSWERData
True/False: The first step in trying to reduce the number of predictors should always be
to use domain knowledge. - ANSWERTrue
True/False: Categorical variables can also be handled by most data mining routines and
do not require extra services. - ANSWERFalse
T/F: The less variables we include, the greater risk of overfitting data - ANSWERFalse
T/F: The goal of SUPERVISED (predictive) learning is to predict a single "target" or
"outcome" - ANSWERTrue
T/F: Regression analysis can be used to develop an equation showing how different
data variables are related. - ANSWERTrue
What are the 3 partitions typically used to address the overfitting problem when building
a predicting model? - ANSWERTraining, validation, and test
A set of measurements on an observation is
• A. Prediction
• B. Profile
• C. Variable
• D. Model - ANSWERProfile
, T/F Predicting the outcome value for new records, given their input values is a
predictive task. - ANSWERTrue
The goal of Supervised Predictive learning is to predict multiple "targets" or "outcome"
variables. - ANSWERFalse
What is the first step while testing for significance T-test?
1. select the test statistic
2. compute the value of the test statistic
3. determine the hypotheses
4. specify the level of significance - ANSWERdetermine the hypotheses
1) What does "e" stand for in the equation Y=B0 + B1*t + e
a) time b) error c) intercept d) trend - ANSWERError
When doing a weighted average the sum of weights must be greater than or equal to
one. - ANSWERFalse
In the Forecast Error Measurement MAD, n stands for a. period number b. actual
demand c. total number of periods d. forecast demand - ANSWERTotal number of
periods
T/F: A simple moving average formula implies equal weighting of prior time periods -
ANSWERTrue
1. (T/F) A short coming of a moving average is that it does not forecast seasonal
component. - ANSWERTrue
T/F: MSE is a measure of the average of the errors absolute value. - ANSWERFalse
T/F: Seasonality accounts for the gradual shifting of the time series over a long period of
time. - ANSWERfalse
T/F- To determine the seasonal factor of a time-series data set, it is important to
develop an average for the same quarters in each period and then find a seasonal
factor by dividing that average by the general average. - ANSWERTrue
T/F: the four components of time series is level, trend, seasonality, and noise. -
ANSWERTrue
The fifth step of decomposition is:
a) Deseasonalize the original data
b) Determine the seasonal factor
c) Create the final forecast by adjusting the regression line by the seasonal factor
