EC 285 - 2026-09-17 - Lecture 03 - Cards

EC 285 - Lecture 03 - Flashcards

30 cards from the lecture notes

Review them with the Spaced Repetition plugin. It keeps your progress inside this file, so the lecture pipeline never rewrites it. Edit or delete any card freely.

#flashcards/EC-285/Lecture-03

Why did Zara Liaqat call the range, the difference between the maximum and the minimum value, a weak measure of dispersion?
?
It is decided entirely by the two outliers and says nothing about how the data are spread between them.

Q1, Q2 and Q3 are the 25th, 50th and 75th percentiles, and Q2 is also the median.

Why does the variance formula square each deviation from the mean instead of just summing the raw deviations?
?
Unsquared deviations above and below the mean cancel out and give a misleadingly small number. Squaring makes every observation count and puts more weight on points far from the mean.

Variance is the average of squared deviations from the average, the phrase Zara Liaqat gave for remembering the formula.

The sample variance is s2=∑i=1n(xi−x¯)2n−1.

Standard deviation and variance come from the identical formula, except that standard deviation takes the square root at the end.

Why do the sample variance and standard deviation formulas divide by n−1 instead of n?
?
Statisticians have proved that n−1 gives a better estimate of the population value, especially when the sample is small.

Why is the standard deviation often more useful than the variance for describing a data set?
?
It is measured in the same units as the original data, which makes it readable.

What follows from the fact that every single observation enters the standard deviation formula?
?
Nothing in the data is ignored, which is the strength of the measure, but outliers inflate the spread it reports.

In statistics notation, μ, σ, σ2 and N describe the population, while x¯, S, s2 and n describe the sample.

The sample coefficient of variation is CV=(Sx¯)⋅100%.

Why can the coefficient of variation compare spread across data sets measured in different units, such as dollars against months?
?
It is a percentage, so it carries no units of its own.

What does the sign of Cov(x,y) tell you about two variables?
?
Positive means they move in the same direction, negative means opposite directions, and zero means they are independent.

The sample correlation coefficient is r=Cov(x,y)sxsy, the covariance standardised by the two standard deviations.

The correlation coefficient always lies between −1 and +1.

In a frequency table of values x with frequencies f, why can you not find the mean by adding the x column and dividing by the number of rows?
?
Each value must be weighted by how often it appears, and the sample size is ∑f, not the number of rows.

For frequency-table data, the sample mean is x¯=∑fixi∑fi.

For frequency-table data, the sample standard deviation is S=∑fi(xi−x¯)2n−1.

In the pairs-of-shoes frequency table, with x running 1 to 7 and frequencies 2, 5, 8, 12, 12, 0, 1, what is the mode?
?
4 and 5, so the data are bimodal. The answer is not 12, because the mode must be in units of x rather than frequency.

The position of the Pth percentile in ranked data is (P100)(n+1), which gives a position and not the answer itself.

In the pairs-of-shoes data, where n=40, how do you arrive at Q1=3?
?
The position is 0.25×41=10.25. Counting cumulative frequencies 2, then 7, then 15, the 10.25th observation falls in the x=3 group.

EC 285 Midterm 1 is on Sunday, October 4, 2026, 12:00 to 1:00 pm, and is worth 20% of the course grade.

EC 285 Midterm 2 is on November 21, 3:00 to 4:00 pm, is worth 20%, and is not cumulative.

The EC 285 final exam is cumulative and worth 35%, or 55% if you drop the lower midterm.

#flashcards/EC-285/Lecture-03 #flashcards/Exam-flagged/EC-285

Which chapters does EC 285 Midterm 1 cover?
?
Chapters 1, 2 and 3 only. Chapter 4 is taught before October 4 but is tested on Midterm 2 instead.

Which EC 285 formulas did Zara Liaqat say will not appear on the exam formula sheet?
?
Variance and standard deviation. She said to memorise them, because there will almost certainly be exam questions on them.

Zara Liaqat flagged this as a likely exam question: why can a correlation of about −0.15 sit alongside an obvious U-shaped pattern in the scatter plot?
?
Correlation measures the strength of a linear relationship only. This pattern is quadratic, so a correlation near zero does not mean there is no relationship.

What did Zara Liaqat say about the covariance and correlation formulas on the EC 285 formula sheet?
?
The formulas will be printed, but the terms in them will not be defined, so you have to know what each symbol stands for.

How far should you round your answers on EC 285 exams?
?
Two decimal places.