QMB3200 Midterm Questions and
Answers
Descriptive Analytics - ANSWER-Describes Past Data
Predictive Analytics - ANSWER-Uses past data to predict Future (inference)
Prescriptive Analytics - ANSWER-Uses models to find Optimal Solution
Scales of Measurement - ANSWER-Nominal & Ordinal (Categorical);
Interval & Ratio (Quantitative)
Nominal Data - ANSWER-Data which consists of names, labels, or categories.
Ordinal Data - ANSWER-Arbitrary scale where the exact numerical value has no significance other than to
rank a set of data points.
Interval Data - ANSWER-Scale in which we know the order and the exact difference between the values.
Ratio Data - ANSWER-Ratios are meaningful. (Length, Width, Weight, Distance)
Categorical - ANSWER-
Descriptive Statistics - ANSWER-Summarizes data with tabular, graphical, and numerical methods
Cross-Sectional Data - ANSWER-Collected at one point in time
Time Series Data - ANSWER-Collected over several time periods
Panel Data - ANSWER-Collected from a group, organized into panels, over time
Frequency Distribution - ANSWER-Tabular summary of data showing the number of observations in each
of several non-overlapping categories or classes
Steps to a Frequency Distribution - ANSWER-1. Determine # of non overlapping classes
2. Determine the width of the classes (largest value - smallest value/# of classes)
3. Determine class limits (each data observation only belongs to 1 class)
Class Midpoint - ANSWER-Halfway between lower and upper class limits
Simpson's Paradox - ANSWER-Reversal of conclusions based on aggregate and unaggregated data
Measures of Location - ANSWER-Mean, Median, Mode
Geometric Mean - ANSWER-Helpful to find average growth of data over time
Measures of Variability - ANSWER-range, variance, standard deviation
Variance - ANSWER-Compares variability of 2 or more variables
Standard Deviation - ANSWER-Simplifies variance by taking square root to make results easier to
compare to other statistics
, Chebyshev's Theorem - ANSWER-Proportion of data within "z" standard deviations from the mean is = to
1 - 1/(z^2)
Coefficient of Variation - ANSWER-Measures how large the standard deviation is relative to the mean
Z Score - ANSWER-Measures location of a value in dataset relative to the mean (in number of standard
deviations)
Any Z-Score Greater than ____ is an Outlier - ANSWER-|3|
Distribution Shapes - ANSWER-Right/Positive Skew: mean > median
Symmetric: mean = median
Left/Negative Skew: mean < median
Why use Chebyshev's Theorem? - ANSWER-Applies to any dataset:
If bell shaped,
68% within 1 deviation of mean
95% within 2 deviations of mean
99.7% within 3 deviations of mean
Covariance - ANSWER-Measures linear association between 2 variables
Positive = Positive Linear Assosication
Correlation Coefficient - ANSWER-Ranges from -1 to +1
1 = positively sloped line (perfect positive linear relationship)
-1 = negatively sloped line (perfect negative linear relationship)
As the data deviates from the slope of the line, it moves closer to 0
Best to show frequency distribution for categorical data - ANSWER-Bar Chart
Best to show relative/percent frequency for categorical data - ANSWER-Pie Chart or Stacked Bar Chart
Best to show frequency distribution for quantitative data - ANSWER-Histogram
Best to show rank order and shape of quantitative data - ANSWER-Stem-And-Leaf
Used to show relationships between 2 quantitative variables - ANSWER-Scatter Diagram
Used to approximate the relationship of data in a scatter diagram - ANSWER-Trendline
Percentiles - ANSWER-
Quartile - ANSWER-Specific percentiles:
Answers
Descriptive Analytics - ANSWER-Describes Past Data
Predictive Analytics - ANSWER-Uses past data to predict Future (inference)
Prescriptive Analytics - ANSWER-Uses models to find Optimal Solution
Scales of Measurement - ANSWER-Nominal & Ordinal (Categorical);
Interval & Ratio (Quantitative)
Nominal Data - ANSWER-Data which consists of names, labels, or categories.
Ordinal Data - ANSWER-Arbitrary scale where the exact numerical value has no significance other than to
rank a set of data points.
Interval Data - ANSWER-Scale in which we know the order and the exact difference between the values.
Ratio Data - ANSWER-Ratios are meaningful. (Length, Width, Weight, Distance)
Categorical - ANSWER-
Descriptive Statistics - ANSWER-Summarizes data with tabular, graphical, and numerical methods
Cross-Sectional Data - ANSWER-Collected at one point in time
Time Series Data - ANSWER-Collected over several time periods
Panel Data - ANSWER-Collected from a group, organized into panels, over time
Frequency Distribution - ANSWER-Tabular summary of data showing the number of observations in each
of several non-overlapping categories or classes
Steps to a Frequency Distribution - ANSWER-1. Determine # of non overlapping classes
2. Determine the width of the classes (largest value - smallest value/# of classes)
3. Determine class limits (each data observation only belongs to 1 class)
Class Midpoint - ANSWER-Halfway between lower and upper class limits
Simpson's Paradox - ANSWER-Reversal of conclusions based on aggregate and unaggregated data
Measures of Location - ANSWER-Mean, Median, Mode
Geometric Mean - ANSWER-Helpful to find average growth of data over time
Measures of Variability - ANSWER-range, variance, standard deviation
Variance - ANSWER-Compares variability of 2 or more variables
Standard Deviation - ANSWER-Simplifies variance by taking square root to make results easier to
compare to other statistics
, Chebyshev's Theorem - ANSWER-Proportion of data within "z" standard deviations from the mean is = to
1 - 1/(z^2)
Coefficient of Variation - ANSWER-Measures how large the standard deviation is relative to the mean
Z Score - ANSWER-Measures location of a value in dataset relative to the mean (in number of standard
deviations)
Any Z-Score Greater than ____ is an Outlier - ANSWER-|3|
Distribution Shapes - ANSWER-Right/Positive Skew: mean > median
Symmetric: mean = median
Left/Negative Skew: mean < median
Why use Chebyshev's Theorem? - ANSWER-Applies to any dataset:
If bell shaped,
68% within 1 deviation of mean
95% within 2 deviations of mean
99.7% within 3 deviations of mean
Covariance - ANSWER-Measures linear association between 2 variables
Positive = Positive Linear Assosication
Correlation Coefficient - ANSWER-Ranges from -1 to +1
1 = positively sloped line (perfect positive linear relationship)
-1 = negatively sloped line (perfect negative linear relationship)
As the data deviates from the slope of the line, it moves closer to 0
Best to show frequency distribution for categorical data - ANSWER-Bar Chart
Best to show relative/percent frequency for categorical data - ANSWER-Pie Chart or Stacked Bar Chart
Best to show frequency distribution for quantitative data - ANSWER-Histogram
Best to show rank order and shape of quantitative data - ANSWER-Stem-And-Leaf
Used to show relationships between 2 quantitative variables - ANSWER-Scatter Diagram
Used to approximate the relationship of data in a scatter diagram - ANSWER-Trendline
Percentiles - ANSWER-
Quartile - ANSWER-Specific percentiles:
