Scribd
Upload
2K views

Edexcel s1 Revision Notes

This document provides information and formulas for statistics concepts including measures of central tendency, skewness, mean, variance, standard deviation, probability, discrete random variables, and their expectations and manipulations. Key points covered are: - Negative skew has mean < median < mode and longer left whisker, positive skew has mode < median < mean and longer right whisker. - Mean is calculated by summing the products of data values and their frequencies and dividing by the total frequency. - Variance is the sum of the squared deviations from the mean divided by the number of data points. Standard deviation is the square root of the variance. - Probability rules include addition rule, multiplication rule, and properties of independent/mutually

Uploaded by

minamakin
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
2K views

Edexcel s1 Revision Notes

This document provides information and formulas for statistics concepts including measures of central tendency, skewness, mean, variance, standard deviation, probability, discrete random variables, and their expectations and manipulations. Key points covered are: - Negative skew has mean < median < mode and longer left whisker, positive skew has mode < median < mean and longer right whisker. - Mean is calculated by summing the products of data values and their frequencies and dividing by the total frequency. - Variance is the sum of the squared deviations from the mean divided by the number of data points. Standard deviation is the square root of the variance. - Probability rules include addition rule, multiplication rule, and properties of independent/mutually

Uploaded by

minamakin
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Statistics 1 - Edexcel

SKEWNESS
Negative Skew
mean < median < mode
(Q2 Q1) > (Q3 Q2)
Longer left hand whisker
Positive Skew
mode < median < mean
(Q2 Q1) < (Q3 Q2)
Longer right hand whisker

Symmetrical
mode = median = mean
(Q2 Q1) = (Q3 Q2)

Equal whiskers

Interpolation to find a median Value =


Lower Class Boundary + (Cumulative Frequency X whatever you want to find, e.g. 0.25 for
Q1, Minus the Cumulative Frequency before the modal class, Divide by Frequency in the
modal class) Then Times by the Class Width.

Mean

The mean is very simple to calculate, If you are given a table with single
integers on the top, e.g. 1,2,3,4(x) , and they have frequencys to them, e.g. the number 1
has a frequency of 5 (f). all you do is make a new table and calculate all the X's multiplied by

the Frequencys, this is FX, then Add them together to get the sum of FX, Efx. Then devide
Efx by the sum of all the Frequencys, EF, which is also the same as CF.
However if the Data is in groups, e.g. 0-9. Then Instead of Summing Efx, You should
Sum the Frequency multiplied by the Midpoints of the Grouped data. e.g. mp of 0-9 is 4.5.
So its Ef(mp) over Ef in grouped data.
Variance and Standard Deviation.
Variance = Sum of X^2 / n (The number of data sets, not the total) - Xbar^2(the mean squared)
or see below for when it involves frequency's
Variance = Sum of FX^2 over the Sum of F, - Xbar^2(The mean squared).
The Standard Deviation is simply the Square Root of the Variance.
Probability
P(Event A or Event B) = P(AuB)
P(Both Events A and B) = P(AnB)

(Remember this one because it had N in the middle like AND)

P(Not Event A) = P(A') or P(Abar)


Complementary Probability, P(A')= 1-P(A)
Addition Rule, P(AuB) = P(A)+P(B) - P(AnB)

(simply add up and minus middle bit)

P(A given B already happened), P(A/B) = P(AnB) / P(B)

(Remember to divide by the whole of B,

its common mistake not to)

Multiplication Rule, P(AnB) = P(A/B) x P(B) or P(AnB) = P(AnB)/P(B) x P(B)


A and B are independent Events if P(AnB) = P(A) x P(B)
A and B are "mutually Exclusive" events if P(AnB) = 0

Regression and Product Moment Correlation Co-efficient


(PMCC)
Sxx = Ex^2 - ((Ex)^2/N)).
Syy= Ey^2-((Ey)^2/n))

(Syy is same format at picture above, change x's to y.)

Sxy = Exy - ((ExEy)/N))

(keep in mind Exy means the total of (X)(Y), so make a new bar on your table

with XY and then add them all up to get Exy)

R = Sxy / (Root(SxxSyy))

The equation of a regression line of y on X is: y=a+bx


B = SXY/SXX

A = Ybar - (B(Xbar)

(Bar just means the average or mean)

To find the equation you first must work out SXY and SXX, Then find B
by dividing SXY by SXX.
Then you find out A by substituting B into the equation A=Ybar-(B(Xbar) Obviously
you will know the mean of Y and X, just Sum of X divided by N. or Sum of Y divided
by N, same as Ex/n or Ey/n
Finally Substitute B and A into the final Equation : Y = A + B(x)
Discrete Random Variables.

You will be given like a table with x on the top and P(X=x) underneath.P(X=x)
means the probability that X takes on a particular value, for instance the
Probability of rolling a 6 with a dice is usually 1/6/

The sum of P(X=x) is always 1. So to find any missing values just add
them all up and equal them to 1 and solve the equation to find missing
number. e.g. 1/6 + 1/2 +1/6 + K = 1, K has to equal 1/6.
E(X) - It is the sum of every value for X multiplied by the probability of getting
that value of X
x P(X=x) xP(X=x)

X^2P(X=x)

0.1

0.1

0.1

0.1

0.3

0.6

4x0.3

0.3

0.9

9x0.3

0.1

0.4

16x0.1

0.1

0.5

25x0.1

E(x) = 0 + 0.1 + 0.6 + 0.9 + 0.4 + 0.5=2.5


E(X^2) = 0 + 0.1 + (4x0.3) + (9x0.3) +(16x0.1) +(25x0.1) (it may help to make a X
squared column) Usually these are in tables and are horizontal. not vertical as shown. sorry
for confusion.
Underneath is another HUGE graph on how to work E(X), i talk about E(X^2) further down.

There is also the Expectation of X^2

E(X^2) = X^2P(X=x)

The reason you work out both Expectation X and X^2 is


because you can now work out the variance of the discrete
random variable. This is calculated by:
Var(X) = E(X^2)- (E(x)^2)
Manipulations of Discrete Random Variables...
E(aX+b) = aE(X)+B
Var(aX+b) = a^2Var(X)
E(2X) = 3E(X)

E(X-2) = E(X) - 2
Var(3X) = 3^2Var(X)
Var(X-2) = Var(X)

it has no effect adding a minus on variance.

Tips to remember all formulas: Rewrite every formula Seven times, its proven writing stuff
out multiple times helps us retain it in our memory. Rewrite more than 7 times if you want
to remember for ever.
i'll be continually updating this up-to the exam on Friday.

You might also like