Use whichever textbook you’d like – the third edition has rock, paper, scissors on the cover and you’d be
able to find used much easier. So that.
Why study stats?
Stats techniques are used to make business decisions
Economists need to test theories
We need to estimate magnitudes of important economic parameters (Marginal propensity to
consume, price elasticity of demand for oil, etc)
He didn’t care about this slide because he said it made it sound like he was going to take over
the world with stats
Methods of organising, summarising and presenting data in an informative way
Frequency distribution, chart forms, central tendency measures, data clustering
“This was this”
Methods used to determine something about a population, based on a sample
Population is the entire set of people or objects of interest or the measurements obtained
A sample is a portion, or part, of the population interest
We use samples so we don’t break everything or spend entirely too much money (i.e. if you
want to figure out how many of a certain car will survive a crash, you don’t drive all of them into
Basically, we make guesses about things based on the data we bring in
“We estimate this was this based on this”
Prohibitive cost of surveying the whole population
Destructive nature of some tests
Physical impossibility of capturing the possibility
Sampling error is why we have this course
Your sample mean won’t be the true one. Your answer won’t be accurate unless you’re lucky
During tests you’re only responsible for things we lecture on
Frequency distribution A grouping of data into mutually exclusive classes showing the number of observations in each
Relative frequency distribution
Class frequencies can be converted to relative class frequencies to show the fraction of the total
number of observations in each class (most people find relative frequencies to be more informative – if
you’re looking at houses, it makes more sense to know that 47% of houses sold were a certain type,
rather than the pure number)
The horizontal axis shows the variable of interest and the vertical axis the amount, number or
fraction of each of the possible outcomes
A distinguishing characteristic of a bar chart is that there is a gap between bars (Histograms
don’t have that, which is the only difference between them. Stats people are anal, apparently)
Can depict any level of measurement – nominal, ordinal, interval, ratio
You can create bar charts in excel – instructions for that are in the textbook if you don’t know
how to do that. Lovely
Decide of the number of classes
Determine class interval or width (must be equal!)
Let’s create intervals of 10% for classes because that’s natural and 1% is stupid.
Set the individual class limits
Tally the list prices into the classes
Count the number of items in each class
(One of the assignments is to look at marks for this class for last year)
Frequently used terms
Class midpoint: point between lower and upper limit
Class interval: subtract the lower limit of the class from the upper limit and get something I
missed in the slide. Yay me
Example of descriptive things`
Report on 96 homes in the south east region. You take the lowest and highest, because they`re
1. Sort data, lowest to highest.
2. Jay and I were talking so I missed this.
3. But it’s in the slides. Sorry Katey and future self.
4. Tally homes
5. Count the tallies. And you have a frequency distribution. (“And you thought stats was going to
be hard. Haha.”) 6. You can make a relative frequency by dividing each frequency by the total number of terms.
(If you then go through and add the first to the second, then the third to that, you get a cumulative
frequency distribution and that tells you what percentage are below a certain level)
He put up a slide of formulas of craziness to say we don`t need to memorise formulas and
during tests he will give us a formula sheet. Yay him.
You try it!
I didn’t want to try.
Relative frequency distribution
We already covered this, so he basically ignored it.
Most people find this more informative. So there’s that.
Similar to the bar chart (doesn’t have gaps)
We’re not going to be fussy about this
Won’t have to do histograms in excel
- Manage options
- Excel toolpack
The next step in summarising data – we took 100 numbers and smashed it into 5 numbers, but
we’re going to get one number here.
Do you want to know how students did in a class? Ask the average.
We are going to learn to spell Chebyshev. That’s what this chapter is. Yay us.
Measures of location