Data skills - probability and distribution

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/17

encourage image

There's no tags or description

Looks like no tags are added yet.

18 Terms

1
New cards

how to find how many value sit above or below a value in R

sum(variable $specific column variable <= number)

  • <= for below this value

  • >= for above this value


2
New cards

how to find how many values sit within a range in R

sum(between(data name$variable, smaller number, larger number)

3
New cards

what is frequency density

The height of a bin or the y-axis variable on a histogram

  • calculated by dividing the frequency of numbers covered in the bin range by the size of the range (class width)

  • class width = highest-lowest value


4
New cards

what are 2 ways to generalise finding to a wider population

  1. Assume the pattern for a smaller sample is the same for the wider population

  • has no real statistical significance

  • can only be used to make a general statement

  1. Assume the data from a sample and population can be represented through a normal distribution

  • provides an overall pattern but can ignore smaller abnormalities


5
New cards

what is normal distribution

a probability distribution that shows that values closer to the mean occur more frequently and are hence more probable

  • A normally distributed histogram will show a unimodal bell curve and follow the empirical rule


6
New cards

How to find the proportion of values above or below a certain number in R

pnorm(q=Num, mean = Number, sd= number)

  • q is the number that values can be above or below

  • replace q with p, and a proportion to calculate max value in that proportion of data


7
New cards

how are standard deviations and mean displayed in normal distribution curve

mean is found at the highest peak of the curve on the x axis. the distribtuion id then split into sections, with each one being mean - or + n standard deviations from the mean

  • If the standard deviation is high, the distribution will be shorter and wider

  • if the standard deviation is low, the distribution will be taller and skinnier


8
New cards

what is the empirical rule

also known as the 68-95-99.7 rule as

  • 68% of data will fall between 1 STD + or - the mean

  • 95% of data will fall between 2 STD + or - the mean

  • 99.7% of data will fall between 3 + or = the mean


9
New cards

how to test in data is normally distributed using R

Using a q-q plot or histogram, but qq plot is easier

  • if the data points follow the line then it can be assumed that the data is normally distributed

  • if they differ from the line too much, it can indicate skewed data


10
New cards

What is a population

an entire group of individuals from which you randomly select individuals from to represent said population

  • larger the sample size the more accurate the results as the more representative the sample


11
New cards

what is a simple random sample

When you are selecting a sample and all individuals are equally likely to be a part of the sample

  • often just inferred but required as assumption for normal distribution


12
New cards

what are parameters

A number calculated from a population which is unknown due to the inability to test said population

  • σ = population standard dev

  • p = population proportion

  • μ = population mean


13
New cards

what are statistic

A number calculated from a sample that can be used to estimate a parameter

  • x̄ = sample mean

  • s = sample standard dev

  • p̂ = proportion of sample


14
New cards

what is needed to estimate unknow parameters from a statistic

to estimate parameters using statistic we need to describe the sampling distribution of values across multiple samples

  • The larger the sample, the closer the statistic gets to the parameter

  • The sampling distribution is calculated using the mean of sample means from multiple experiments ( is an estimate of the population mean)


15
New cards

how to determine the standard error of a sample mean

n = sample size

σx = sample mean

σ= population standard dev

  • increase in sample size leads to decrease in standard deviation


<p>n = sample size</p><p>σx = sample mean</p><p>σ= population standard dev</p><ul><li><p>increase in sample size leads to decrease in standard deviation</p></li></ul><p></p>
16
New cards

how to calculate the normal distribution of a simple random sample

sample mean = n(population mean, population standard dev/the square root of n

17
New cards

what is the central limit theorem

The sample distribution of a sample mean approaches a normal distribution the larger the sample size is

  • A general rule of thumb is that is the sample size is over 30 then we can assume a normal distribution of our sampling distribution


18
New cards

how to figure out that impossible quiz question

For a random variable (x) normally distributed with mean and standard deviation, each value of n will be a certain number of standard deviations way from the mean

  • if the curve seems to sit one unit away from the mean, it means that most of the data sit 3 STD according to the empirical rule