Quiz
The normal distribution
Dieses Quiz enthält Fragetypen, die sich im gewohnten Player öffnen.
Fragen in diesem Quiz (11)
1the [blank_start]normal[blank_end] distribution occurs very [blank_start]often[blank_end] in nature > therefore, is a very important distribution for the statistical analysis of data
- it is characterised by a symmetrical "[blank_start]bell[blank_end] shape" curve > which corresponds to values near the [blank_start]mean[blank_end] being more common, which values further away from the mean are [blank_start]less[blank_end] common, "tail off" in terms of their frequencies- Anormal
- Boften
- Cbell
- Dmean
- Eless
2in order to understand what it really means for the data to be [blank_start]normally[blank_end] distributed, we first need to consider the idea of probability
[blank_start]probability[blank_end] = is used to measure the likelihood of an event happening
if we were to repeat an experiment over and over again, the probability of a particular outcome A is defined as the proportion of the total number of repeats in which A would actually occur, if we were to keep on repeating the experiment we denote this probability by [blank_start]Pr(A)[blank_end]
example 1: rolling a fair dice
Let event A be the dice lands with three spots face up; Pr(A) = 1/[blank_start]6[blank_end] = 0.167
note that in this experiment there are six equally likely outcomes, all with probability of [blank_start]1[blank_end]/6
note that the probability scale runs between 0 and [blank_start]1[blank_end] inclusive. The higher the number, the [blank_start]more[blank_end] likely the event will be, the sum of the probabilities of all possible events must equal [blank_start]1[blank_end]- Anormally
- Bprobability
- CPr(A)
- D6
- E1
- F1
- Gmore
- H1
3probability example 2: what is probability of a randomly chosen individual having fair hair or blue eyes or both
- we can estimate the probability of [blank_start]blue[blank_end] eyes by looking at the proportion in a sample with blue eyes
- we can similarly estimate the probability with [blank_start]fair[blank_end] hair and both fair hair and blue eyes
TO CALCULATE THIS
Pr(fair hair or blue eyes or both) = Pr(fair hair) + Pr(blue eyes) [blank_start]-[blank_end] Pr(fair hair AND blue eyes)
this is an example of the [blank_start]addition[blank_end] rule of probability; the minus sign is to avoid double counting- Ablue
- Bfair
- C-
- Daddition
4probability example 3: what is the probability of a randomly chosen male having fair hair and being at least 6 foot tall
- we can estimate the probability by looking at the proportion of males who are at least [blank_start]6[blank_end] foot tall and have fair hair
- HOWEVER, we can assume that hair colour is unrelated to height then so in this case (BECAUSE THESE 2 VARIABLES/CHARACTERISTICS ARE [blank_start]INDEPENDENT[blank_end], hair colour does not effect height)
pr(fair hair and height ovr 6ft) = P(fair hair)[blank_start]x[blank_end]Pr(height over 6ft)
this is called the [blank_start]multiplication[blank_end] rule of probability; this would [blank_start]not[blank_end] work if the two factors are related (not independent)- A6
- BINDEPENDENT
- Cx
- Dmultiplication
- Enot
5probability measurements only really make sense when for discrete outcomes i.e. when we can make a list of [blank_start]all[blank_end] the possible outcomes
when the measurements are on a [blank_start]continuous[blank_end] scale, such as the length of animals, then there are infinitely many possible outcomes, and it is [blank_start]not[blank_end] possible to list them
this is where the [blank_start]normal[blank_end] distribution can help us:
the distributions of lobster lengths have roughly the normal distribution shape for each sex- Aall
- Bcontinuous
- Cnot
- Dnormal
6[blank_start]Minitab[blank_end] can fit the normal curves for us:
- suppose we were to take a new measurement for a different female lobster denoted by the outcome X
> there would be infinite possibilities for the length could be as specific as mm, so it doesn't make sense to ask about the probability of a specific outcome
instead, meaningful questions will ask if x lies in a particular [blank_start]range[blank_end], e.g. what is the probability that X will be less than 165?
- AMinitab
- Brange
7calculating the area under the curve (our particular range of interest)
- the probability of X lying in any particular range of numbers is given by the [blank_start]area[blank_end] under the curve for the range of interest, the total area under the curve being equal to [blank_start]1[blank_end]
the [blank_start]normal[blank_end] distribution curve is given by a mathematical formula (image)
- the distribution is completely determined by 2 parameters, the [blank_start]mean[blank_end] and the standard [blank_start]deviation[blank_end]
this means it is possible to calculate areas under the curve, and hence probabilities
- fortunately minitab can do this for us- Aarea
- B1
- Cnormal
- Dmean
- Edeviation
8the [blank_start]normal[blank_end] distribution is useful for it is then used to model our data, and makes it easy to calculate the probability of an outcome in any given range (i.e. below/above a value)
- can only do this because we our modelling with a normal distribution
- the normal distribution acts as a model for the way data behaves- Anormal
9An important property of the normal distribution is that [blank_start]95[blank_end]% of our data is going to lie within [blank_start]1.96[blank_end] (we say 2) standard deviations of the mean
similarly, [blank_start]two[blank_end]-thirds of the data lies within 1 standard [blank_start]deviation[blank_end] of the mean and virtually all the data lies within [blank_start]3[blank_end] standard deviations of the mean- A95
- B1.96
- Ctwo
- Ddeviation
- E3
10The distribution of the sample mean
- looking at IQ, it is measured on a scale which is designed so that population distribution is normal, with mean [blank_start]100[blank_end] and standard deviation 15- A100
11[blank_start]Central[blank_end] limit theorem
- suppose we draw a sample of n independent (un-related) observations from a population with mean and standard deviations
if we have a big sample, the distribution of the same mean is, to a good approximation:
standard [blank_start]deviation[blank_end] gets tighter, decreases with increasing sample size
- the approximation gets better and better as the sample size [blank_start]increases[blank_end]
- this is true whatever the original distribution of the data (does not need to be a normal distribution)- ACentral
- Bdeviation
- Cincreases
Informationen
- Beschreibung
- Stats Quiz am The normal distribution, erstellt von sophie waddington am 24/12/2018.
- Autor
- sophie waddington
- Sprache
- English
- Aufrufe
- 0
- Schlagwörter
- Bewertung
- Noch keine Bewertungen Bewerte dieses Quiz