Data Analytics - Unit-3
Aa dependent variable and one or more independent variables
Btwo databases
Ctwo networks
Dtwo programs
Correct Answer
a dependent variable and one or more independent variables
ABasic Linear Unified Equation
BBest Logical Unbiased Error
CBest Linear Unbiased Estimator
DBinary Linear Uniform Estimate
Correct Answer
Best Linear Unbiased Estimator
Athe data has no variables
Bthe sample has only one value
Cthe classical regression assumptions hold
Dthe errors are always zero
Correct Answer
the classical regression assumptions hold
Athe number of variables
Bthe sum of squared residuals
Cthe sum of the observations
Dthe mean of the data
Correct Answer
the sum of squared residuals
Athe difference between the observed and the predicted value
Bthe mean of Y
Cthe slope of the line
Dthe number of observations
Correct Answer
the difference between the observed and the predicted value
Athe proportion of variance in the dependent variable explained by the model
Bthe number of outliers
Cthe number of predictors only
Dthe mean of residuals
Correct Answer
the proportion of variance in the dependent variable explained by the model
Aindependent variables are highly correlated with each other
Bthe dependent variable is constant
Cthere is only one predictor
Dresiduals are zero
Correct Answer
independent variables are highly correlated with each other
AGini index
BSilhouette score
CVariance Inflation Factor (VIF)
DROC curve
Correct Answer
Variance Inflation Factor (VIF)
Aerrors are always positive
Bmean of Y is zero
Cvariance of the errors is constant
Dvariables are categorical
Correct Answer
variance of the errors is constant
Amissing values
Bmulticollinearity only
Cautocorrelation in the residuals
Doutliers only
Correct Answer
autocorrelation in the residuals
AY = X / Ξ²
BY = Ξ²0 + Ξ²1X + Ξ΅
CY = Ξ²0 Γ Ξ²1
DY = Ξ΅ only
Correct Answer
Y = Ξ²0 + Ξ²1X + Ξ΅
Athe value of Y when X is large
Bthe error term
Cthe change in Y for a one-unit change in X
Dthe number of observations
Correct Answer
the change in Y for a one-unit change in X
Apenalises the addition of irrelevant predictors
Bis always larger than R-squared
Cignores the number of predictors
Dcan never be used for model comparison
Correct Answer
penalises the addition of irrelevant predictors
Aadding random variables
Bdeleting all predictors
Cduplicating the data
Dselecting a meaningful subset of variables for the model
Correct Answer
selecting a meaningful subset of variables for the model
ABootstrap selection
BPixel selection
CRandom pruning
DForward selection
Correct Answer
Forward selection
Acontinuous only
Balways a date
Calways an image
Dbinary or categorical
Correct Answer
binary or categorical
A-1 and 0
B0 and 1
C0 and 100
D1 and 10
Correct Answer
0 and 1
Ae^z only
BzΒ²
Clog(z)
D1 / (1 + e^(βz))
Correct Answer
1 / (1 + e^(βz))
Ap Γ (1 β p)
Bp / (1 β p)
C1 / p
DpΒ²
Correct Answer
p / (1 β p)
Athe natural logarithm of the odds
Bthe square of the odds
Cthe probability itself
Dthe residual
Correct Answer
the natural logarithm of the odds
Aordinary least squares only
Brandom guessing
Cmaximum likelihood estimation
Dthe median
Correct Answer
maximum likelihood estimation
ADurbinβWatson test
BShapiroβWilk test
CKruskal test
DHosmerβLemeshow test
Correct Answer
HosmerβLemeshow test
Aprecision against time
Btrue positive rate against false positive rate
Cresiduals against X
Dmean against variance
Correct Answer
true positive rate against false positive rate
Ais no better than random guessing
Bis perfect
Calways predicts correctly
Dhas no variables
Correct Answer
is no better than random guessing
ACredit scoring and customer churn prediction
BPredicting the exact temperature
CCompressing images
DSorting an array
Correct Answer
Credit scoring and customer churn prediction
Fill in the Blanks
26
BLUE stands for Best Linear __________ Estimator.
Correct Answer
Unbiased
27
The least squares method minimises the sum of __________ residuals.
Correct Answer
squared
28
__________ is the proportion of variance in the dependent variable explained by the regression model.
Correct Answer
R-squared
29
High correlation among the independent variables is called __________.
Correct Answer
multicollinearity
30
The __________ Inflation Factor (VIF) is used to detect multicollinearity.
Correct Answer
Variance
31
Constant variance of the error terms is called __________.
Correct Answer
homoscedasticity
32
The DurbinβWatson test checks the residuals for __________.
Correct Answer
autocorrelation
33
Logistic regression predicts a __________ outcome such as yes or no.
Correct Answer
binary
34
The logistic function is also called the __________ function.
Correct Answer
sigmoid
35
The odds of an event with probability p are p / __________.
Correct Answer
1 β p
36
The natural logarithm of the odds is called the __________.
Correct Answer
logit
37
Logistic regression coefficients are estimated by __________ likelihood estimation.
Correct Answer
maximum
38
The ROC curve plots the true positive rate against the __________ positive rate.
Correct Answer
false
39
An AUC of __________ indicates a model that is no better than random guessing.
Correct Answer
0.5
40
Forward selection and backward elimination are types of __________ regression.
Correct Answer
stepwise