Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
55 Cards in this Set
- Front
- Back
List of types of Summary measures
|
1. Measure of Location
2. Measure of Scatter 3. Shape of data |
|
Measures of Location are also know as ...
|
Measures of central tendency.
|
|
What is the purpose of the Measures of Location
|
To show, in general terms, the size of the data.
|
|
What is the arithmetic mean?
|
Sum of readings divided by the number of readings
|
|
What is the median?
|
The middle value of a set of numbers. There is no mathmatical formula.
|
|
How do you determine median if there is an even number of values?
|
Take the arithmetic mean of the middle two values as the median.
|
|
What is the Mode?
|
The most frequently occurring value.
|
|
How are measures of location calculated for grouped or classified data (frequency table)?
|
Each class is represented by its mid-point, which is used for calculating measures of location.
|
|
How is the median calculated from a frequency table?
|
Use the mid-point of the class in which the middle value occurs.
|
|
How is the mode determined from a frequency table?
|
Use the mid-point of the class with the highest number of observations.
|
|
Most meaningful measure of location for: symmetrical, u-shaped and j-shaped distributions?
|
Symmetrical - arithmetic mean
U Shape - modes (2) J Shape - median |
|
What are other uses for Measures of Location?
|
1. Focus for the eye
2. For comparison |
|
What are good reasons not to use the Arithmetic Mean as the Measure of Location?
|
1. Distortion of the mean by outliers
2. Distortion of the mean by clusters 3. Error by taking averages of averages |
|
What alternatives should be considered for:
a. Distortion by outliers b. Distortion by clusters c. Error by average of average |
a. The median should be considered
b. The mode should be considered c. Return to definition of average and correct the calculation |
|
Define Measures of Scatter.
|
They are a measure of the extent to which the readings are grouped closely together or are scattered over a wide interval.
|
|
Measures of Scatter are also known as…
|
Measures of dispersion
|
|
List the Measures of Scatter.
|
1. Range
2. Interquartile Range 3. Mean Absolute Deviation 4. Variance 5. Standard Deviation |
|
Explain Range
|
total interval covered by the numbers
|
|
Explain Interquartile Range
|
The range of the numbers after having eliminated the highest and lowest 25%.
|
|
Mean Absolute Deviation
|
is the sum of the absolute difference between each reading and mean divided by the number of readings
|
|
Variance
|
Sum of squares of deviations of each reading from the mean divided by the number of readings minus 1.
|
|
Standard Deviation
|
The square root of the Variance
|
|
Calculation of measures of dispersion
|
1. Calculator
2. Spreadsheet 3. Shortcut formula for calculating Variance |
|
Pros and cons of Range
|
Advantage - Easily understood and familiar
Disadvantage - Distorted by outliers; descriptive only |
|
Pros and cons of Interquartile Range
|
Advantage - Easily understood
Disadvantage - Not well known; descriptive only |
|
Pros and cons of Mean absolute deviation
|
Advantage - Intuitively sensible
Disadvantage - Unfamiliar; mathematically difficult |
|
Pros and cons of Variance
|
Advantage - Easy math; used in other theories
Disadvantage - Wrong units; No intuitive appeal |
|
Standard Deviation
|
Advantage - Easy math; used in other theories
Disadvantage - Too involved for descriptive purposes |
|
How is the method of calculation for the measure of scatter selected?
|
If for purely descriptive - MAD is probably best.
If measure of scatter needed for wider financial or statistical theory, then variance and standard deviation commonly used. |
|
Define Coefficient of Variation
|
Standardizes the standard deviation so that comparisons can be made.
|
|
Formula for Coefficient of Variation
|
Standard deviation divided by the arithmetic mean.
|
|
What other Summary Measures are occasionally encountered?
|
Skew and Kurtosis
|
|
Define skew
|
measures the extent to which a distribution is non-symmetrical
|
|
Explain:
Left Skew Zero Skew Right Skew |
Left Skew (Negatively skewed) - hump is leaning to the right.
Zero Skew - distribution is symmetrical Right Skew (Positively skewed) - hump is leaning to the left. |
|
Define Kurtosis
|
A generally qualitative approach to the degree to which the distribution is 'punched in' or 'filled out'
|
|
Explain shape of;
low kurtosis medium kurtosis high kurtosis |
Low kurtosis - upside down funnel
Medium kurtosis - upside down 'V' High kurtosis - upside down 'U' |
|
Options in dealing with Outliers
|
1. Twyman's Law - data that looks unusual or interesting is wrong. Check for typo or other error in data.
2. Part of the pattern - should not change or exclude 3. Isolated event - not error, but unlikely to be repeated (not part of the pattern) - usually exclude |
|
Procedure for outliers
|
1. Look for mistakes and correct them
2. Decide if part of the pattern or not - to include or exclude. |
|
What is an Index?
|
a measure used to summarize the movement of a variable over time. Makes numbers easier to understand and compare with other series.
|
|
Best known index
|
Cost of Living Index - cost of many different goods, fuel, transport, etc.
|
|
What is the index base year?
|
The starting point for the series, when the index was 100.
|
|
Define Simple Index
|
Result of the conversion of one series of numbers into another based on 100
|
|
What are limits for index
|
A factor of three before and after the base year.
|
|
What might happen to the base year in a long series?
|
There may be multiple base years.
|
|
Define a Simple Aggregate Index
|
An index which is used to summarize several factors.
|
|
Define a price-relative index
|
to counter the impact that higher numbers have on index versus the lower.
|
|
Define a Weighted Aggregate Index
|
An aggregate index in which you can assign different weightings to reflect the importance of the various components.
|
|
Define a Laspeyres Index
|
A Weighted Aggregate Index in which the base month quantities are used for the weighting.
|
|
What is the major criticism of the Laspeyres Index?
|
The weightings used in the base year may over time become un representative.
|
|
Alternative to Laspeyres Index.
|
Paasche Index
|
|
Define the Paasche Index
|
Instead of using base year quantities as weightings, they use the quantities from the most recent month. However, everytime the weightings change, the entire series must be recalculated.
|
|
Define a fixed weight index
|
weights are from neither the base period nor the most recent. They are from an intermediate period, or from the average of several periods.
|
|
How and why is economic data sometimes deflated (e.g. GNP)?
|
To deflate the GNP is to manipulate the figures to remove the effect of increasing prices on GNP. Makes seeing the true increase/decrease easier. This is done by dividing the current GNP by the current year Price Index and multiplying by 100. This results in the GNP (real).
|
|
What are synonyms for model?
|
Pattern or summary
|
|
What information should be included in the model?
|
1. Number of readings
2. A measure of location 3. A measure of scatter 4. The shape of the distribution |