Analytics Exam 2

Simple linear regression describes a linear relationship between _____

a. one independent variable and one dependent variable
b. two or more independent variables
c. two dependent variables and one independent variable
d. two or more dependent variables

A

In simple linear regression, the independent variable is typically represented by _____.

a. Y
b. X
c. R
d. β

B

In simple linear regression, the dependent variable is represented by _____.

a. X
b. R
c. Y
d. β

C

Multiple regression involves ____

a. one independent variable and one dependent variable
b. two or more dependent variables
c. two dependent variables and one independent variable
d. two or more independent variables

D

Predicting blood pressure from variables such as height, weight, age, and exercise is an example of ____ regression.

a. multiple
b. simple
c. binary
d. random

A

In multiple regression, Multiple R represents the _____.

a. regression slope
b. multiple correlation coefficient
c. sample size
d. coefficient of determination

B

In multiple regression, R Square represents the _____.

a. correlation intercept
b. multiple correlation coefficient
c. coefficient of determination
d. independent variable

C

R-Squared (R²) measures the ____ of the line to the data.

a. slope
b. distance
c. variance
d. fit

D

The value of R-Squared lies between _____.

a. 0 and 1
b. −1 and 1
c. 0 and 100
d. 1 and 10

A

An R-Squared value of 1 indicates _____.

a. no relationship
b. a perfect fit
c. weak correlation
d. random variation

B

The larger the R-Squared value, the ____ the fit of the regression line.

a. weaker
b. earlier
c. better
d. smaller

C

The regression summary output provides the ____ for independent variables & intercept, which is used to test for significance.

a. R-values
b. slopes
c. intercepts
d. p-values

D

The significance level (α) is defined as ____
a. 1 - the confidence level
b. 1 + the confidence level
c. 2 x the confidence level
d. P value x confidence level

A

For an independent variable, if p-value is less than or equal to significance level (α),
a. The independent variable is not significant
b. The independent variable is significant
c. The dependent variable exceeds the independent variable
d. The independent variable is inaccurate

B

If confidence is 95%, what is the significance level (α)?
a. 95%
b. 105%
c. 5%
d. Not enough information

C; 1 - 0.95 = 0.05 or 5%

In regression analysis, insignificant independent variables are typically ____ the model.

a. added to
b. adjusted to
c. ignored in
d. removed from

D

P-values are used to check the ____ of independent variables in regression models.

a. slope
b. significance
c. variance
d. sample size

B

The lower the p-value, the ____ the statistical significance.

a. weaker
b. greater
c. smaller
d. earlier

B

P-values less than or equal to 0.05 are considered ____ significant.

a. rarely
b. statistically
c. partially
d. occasionally

B

Given the following P-Values, which of the following is the least significant?
a. 0.013
b. 1.011E-10
c. 0.456
d. 0.0005

C; The largest p-value is the least significant

In the systematic model building approach, the first step is to construct a model with:
a. only significant independent variables
b. one independent variable
c. all available independent variables
d. only the dependent variable

C

Significance of the independent variables is checked by examining the _____.
a. adjusted R²
b. observations
c. confidence level
d. p-values

D

In step 2 of the systematic model building approach, identify the independent variable having the largest _____ that exceeds the chosen level of significance α.
a. p-value
b. R Square
c. coefficient
d. observation

A

In step 3 of the systematic model building approach after identifying the largest variable, remove it from the model and evaluate _____.
a. Multiple R
b. Adjusted R²
c. Standard Error
d. the intercept

B

When removing variables from the systematic model building approach, remove ____ at a time.
a. all variables
b. two variables
c. one variable
d. no variables

C

When removing variables from the systematic model building approach, continue removing the largest p-values until all variables are _____.
a. dependent
b. removed
c. equal
d. significant

D

In regression analysis, an interaction occurs when:
a. the effect of one variable is dependent on another variable
b. independent variables are removed from the model
c. the regression equation contains only one predictor
d. the dependent variable equals zero

A

In regression analysis, multicollinearity occurs when
a. the dependent variable has no variation among the independent variables
b. there are strong correlations among the independent variables
c. the regression line perfectly fits the data
d. only one independent variable is used

B

Good regression models should include only _____ independent variables.
a. dependent
b. random
c. significant
d. categorical

C

The p-value is used to evaluate the significance, and _____ means it is significant.
a. > .05
b. = 1
c. < 1
d. < .05

D

If independent variables in a regression analysis are not significant, don’t remove them _____.
a. all at once
b. from the dataset
c. by p-value
d. from the model

A

If independent variables are not significant, different regression models may result in different _____.
a. dependent variables
b. significances
c. sample sizes
d. observations

B

Adding an independent variable to a regression model will always result in an ____ than the original.
a. Equal or less than p-value than the original
b. Equal or higher standard error than the original
c. Equal or higher R² than the original
d. α of 0.

C

Adjusted R² reflects both the number of independent variables and the _____ size.
a. effect
b. model
c. variable
d. sample

D

An increase in the Adjusted R² indicates the regression model has _____.
a. improved
b. failed
c. repeated
d. narrowed

A

Good regression models have the fewest number of explanatory variables providing an adequate interpretation of the dependent variable. This is called _____.
a. significance
b. parsimony
c. interaction
d. multicollinearity

B

For a simple linear regression model, significance of regression is
a. a test to determine whether the sample size is large enough for the model
b. a calculation used to estimate the slope of the regression line
c. a hypothesis test of whether the true regression coefficient β₁ is zero
d. a method used to determine the value of the dependent variable

C

There are _____ major categories of forecasting approaches.
a. two
b. four
c. five
d. three

D

What are the 3 major categories of forecasting approaches?

1. Qualitative and judgemental techniques

2. Statistical time-series models

3. Explanatory/casual models

Qualitative and judgmental forecasting relies on _____.
a. significance and forecasting
b. regression and experience
c. comparison and regression
d. experience and intuition

D

Qualitative and judgmental forecasting is used when (2)
a. historical data is not available
b. predictions are needed for the future
c. random numbers are generated
d. Historcal data is sampled from the population

A and B

The _____ is a qualitative/judgemental analysis that obtains a forecast through comparative analysis with prior situations.
a. Prior method
b. Sample method
c. Historical analogy method
d. Delphi method

C

The _____ is a qualitative/judgemental analysis that questions an anonymous panel of experts multiple times.
a. Prior method
b. Historical analogy method
c. Sample method
d. Delphi method

D

A time series model is a stream of
a. time-based categorical data
b. historical time-series data
c. opinion-based data
d. experiment data

B

Weekly sales is an example of:
a. Ordinary data
b. Categorical data
c. Time series data
d. Other data

C

Time series models are best for _____ forecasting problems.
a. long-range
b. judgmental
c. short-range
d. causal

C

A seasonal effect is one that _____
a. Shows ups and downs over several years
b. usually fluctuates in short time scales, like days
c. has no or minimal pattern
d. repeats at fixed intervals of time, like weeks, months, or days

D

Cyclical effects describe ____
a. ups and downs over several years
b. fluctuations that occur within a day
c. repeated fluctuations multiple times a year
d. completely random variation that cannot be explained

A

Indicators are _____, like unemployment rate on economic trends
a. values that replace the original data with new predictions and results
b. measures that are believed to influence the behavior of a variable
c. predictions made after observing the results of past data
d. temporary measurements used only for seasonal adjustments

B

Indicators are often combined quantitatively into an index, a single measure that weights multiple _____.
a. variables
b. forecasts
c. indicators
d. models

C

A(n) ____ is a single measure that weighs multiple indicators, thus providing a measure of overall expectation
a. Indicator
b. Variable
c. Trend
d. Index

D

An index is used in _____ forecasting.
a. judgmental
b. statistical
c. random
d. cyclical

A

T/F Indexes provide direction of change but not a complete forecast.

True

A value randomly generated from a specified probability distribution is called a(n) _____.
a. parameter
b. random variate
c. estimate
d. sample population

B

Triangular distributions are used in the absence of data and depend on which parameters?
a. mean and variance
b. lower bound and upper bound
c. minimum, maximum, and most likely values
d. minimum and maximum

C

Uniform distributions are used in the absence of data and depend on which parameters?
a. mean and standard deviation
b. lower bound and upper bound of the sample
c. minimum, maximum, and most likely values
d. minimum and maximum

D

What is the formula for uniform distribution (U)?

U = a + (b - a)*RAND()

Consider the following formula: U = a + (b - a)*RAND()

If a = 20 and B = 25, what is the purpose of this uniform formula?

It generates random numbers between 20 and 25.

NORM.INV and NORM.S.INV are examples of
a. inverse functions that generate random numbers
b. functions that help with regression forecasting
c. inverse functions that formulate probability distributions
d. time-series index formulas

A

A ____ is a tool for building mathematical and statistical models that characterize relationships between a dependent variable and one or more independent/explanatory/ratio/categorical variables.
a. forecasting model
b. regression analysis
c. time-series index
d. probability distribution

B

In a regression analysis, explanatory variables may be ratio or categorical, but all of them are _____.
a. dependent
b. random
c. seasonal
d. numerical

D

Ratio variables are interval variables but with the added condition that ______.
a. 0 indicates there is none of that variable
b. values can be negative or positive
c. the units between values are unequal
d. the variable must be categorical

A

Is temperature a ratio variable?

No; there can’t be no temperature.

Is weight a ratio variable?

Yes; there can be no weight

W

A

In regression models of time-series data, the independent variable is _____.
a. the dependent variable
b. time
c. correlation
d. slope

B

In regression models of time-series data, the focus is on
a. analyzing relationships
b. explaining variation
c. predicting the future
d. removing insignificant factors

C

A regression model that involves a single independent variable is known as _____ regression.
a. multiple
b. time-series
c. simple
d. causal

C

A regression model that involves two or more independent variables is called _____ regression.
a. simple
b. ratio
c. explanatory
d. multiple

D

What is the purpose of a trendline tool?

a. Shows the best fitting functional relationship for a set of data
b. Calculates the probability distribution of the data
c. Measures the variability of residual errors
d. Determines the number of independent variables

A

Multiple R is another name for the sample _____ coefficient.

a. regression
b. determination
c. correlation
d. variance

C

The value of Multiple R varies between _____.

a. 0 and 1
b. −1 and +1
c. −1 and 0
d. 0 and 10

B

A value of 1 for Multiple R indicates a _____ relationship.

a. perfect negative
b. zero
c. perfect positive
d. random

C

A value of 0 for Multiple R indicates _____ relationship.

a. a weak
b. a strong
c. a negative
d. no

D

Adjusted R² adjusts the value of R² based on the number of _____ variables.

a. dependent
b. explanatory
c. independent
d. categorical

C

Adjusted R² tells how much variation in the _____ variable is accounted for by the model.

a. independent
b. dependent
c. categorical
d. random

B

_____ measures the variability of the observed Y values from the predicted Y values.

a. Standard Error
b. Multiple R
c. Adjusted R²
d. Correlation coefficient

A

The standard error is formally called the Standard Error of the _____.

a. regression
b. estimate
c. equation
d. prediction

B

When the data points are clustered close to the regression line, the standard error will be _____.

a. larger
b. random
c. small
d. negative

C

Generally, the more scattered the data points are from the regression line, the _____ the standard error.

a. smaller
b. weaker
c. lower
d. larger

D

____ are the differences between the actual values and the predicted values of the dependent variable.

a. Observations
b. Estimates
c. Residuals
d. Errors

C

Observed errors can be _____.

a. positive or negative
b. only positive
c. only negative
d. equal to zero

A

Standard residuals describe

a. the predicted values of the dependent variable produced by the regression equation
b. how far each residual is from the mean of the data in units of standard deviation
c. the difference between the observed and predicted values of the dependent variable
d. the slope coefficient that measures the relationship between independent variables

B

An ____ is a value that is different from the rest of the data.

a. observation
b. average
c. error
d. outlier

D

Excel’s Regression Tool can be used for both _____ and ______ linear regressions.

a. simple; multiple
b. categorical; binary
c. time-series; multiple
d. categoritcal; simple

A

The regression tool provides outputs such as _____, _____, and _____.

R², Adjusted R², Multiple R

In the regression tool, the independent variables must be in _____ columns.

a. separate
b. random
c. contiguous
d. labeled

C

In Excel regression, the intercept can be forced to zero by selecting the _____ option.

a. zero constant
b. remove intercept
c. intercept adjustment
d. constant is zero

D

Regression analysis assumes that the errors are _____ distributed with a mean of _____.

a. uniformly; one
b. normally; zero
c. randomly; one
d. exponentially; zero

B

Normality of residuals can be checked using a _____.

a. scatterplot
b. regression equation
c. histogram
d. correlation matrix

C

T/F It is usually easy to evaluate normality with small sample sizes. Example: Seeing what is normal based on 5 records.

False; it is usually difficult.

If you only have 5 records, it’s hard to say if that’s normal because it is a small sample size.

Cross-sectional and time-series models are all examples of ___ models.
a. Contiguous
b. Multiple R
c. Regression
d. Linear

C

This regression model is used for forecasting the future
a. Cross sectional
b. Time series

B

This regression model is carried out at a single point in time on a statistical unit
a. Cross sectional
b. Time series

A

In time-series regression models, time (or function of time) is the ____ variable
a. Dependent
b. Contiguous
c. Random
d. Independent

D

If a regression model has 1 independent variable it is a ____ regression.
a. multiple linear
b. cross-sectional
c. simple
d. time-series

C

If a regression model has 2 or more independent variables it is a _____ regression
a. simple linear
b. time-series
c. cross-sectional
d. multiple linear

D

An influence diagram is:
a. a visual representation of a descriptive model that shows how the elements relate
b. a mathematical model used to calculate optimal solutions for complex decision problems
c. a statistical method used to measure relationships between dependent and independent variables
d. a forecasting tool used to predict future outcomes based on historical time-series data

A

Demand influences profit by:
a. determining the price customers are willing to pay for a product
b. predicting how many units of a product will be sold
c. measuring the variability of costs associated with production
d. identifying the relationship between sales and marketing expenses

B

Quantity produced is
a. a production level determined by available resources
b. a forecast value used to estimate future sales
c. a decision option typically based on demand
d. a quantity measured after production is completed

C