AQA Further Maths A Level

Statistics
Formula Sheet

Provided in formula book Not provided in formula book

This work by PMT Education is licensed under https://bit.ly/pmt-cc

https://bit.ly/pmt-edu-cc CC BY-NC-ND 4.0

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Discrete Random Variables and Expectation

Measures of Average and Spread for a Discrete Random Variable

Expectation 𝐸𝐸(𝑋𝑋) = � 𝑥𝑥𝑖𝑖 𝑝𝑝𝑖𝑖

𝑖𝑖

2
𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) = � 𝑥𝑥𝑖𝑖2 𝑝𝑝𝑖𝑖 − �𝐸𝐸(𝑋𝑋)�
Variance 𝑖𝑖
= 𝐸𝐸(𝑋𝑋 2 ) − (𝐸𝐸(𝑋𝑋))2

Standard Deviation 𝜎𝜎 = �𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋)

The value of 𝑋𝑋 which has the

Mode
largest probability

𝑃𝑃(𝑋𝑋 ≤ 𝑀𝑀) = 0.5

Median
𝑃𝑃(𝑋𝑋 ≥ 𝑀𝑀) = 0.5

Functions of Discrete Random Variables

𝐸𝐸(𝑎𝑎𝑎𝑎 + 𝑏𝑏) = 𝑎𝑎𝑎𝑎(𝑋𝑋) + 𝑏𝑏

𝑉𝑉𝑉𝑉𝑉𝑉(𝑎𝑎𝑎𝑎 + 𝑏𝑏) = 𝑎𝑎2 𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋)

Expectation of a General Function of a Discrete Random Variable

𝐸𝐸(𝑔𝑔(𝑋𝑋)) = � 𝑔𝑔(𝑥𝑥𝑖𝑖 )𝑝𝑝𝑖𝑖

Discrete Uniform Distribution

1
𝑃𝑃(𝑋𝑋 = 𝑥𝑥) = 𝑛𝑛 for 𝑥𝑥 = 1,2, … 𝑛𝑛

𝑛𝑛 + 1
𝑋𝑋~𝑈𝑈(𝑛𝑛) 𝐸𝐸(𝑋𝑋) =
2
𝑛𝑛2 − 1
𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) =
12

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Poisson Distribution

−𝜆𝜆
𝜆𝜆𝑥𝑥
𝑃𝑃(𝑋𝑋 = 𝑥𝑥) = 𝑒𝑒
𝑥𝑥!

𝑋𝑋~𝑃𝑃𝑃𝑃(𝜆𝜆) 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 = 𝜆𝜆

𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 = 𝜆𝜆

Sum of Independent Poisson Distribution

When 𝑋𝑋~𝑃𝑃𝑃𝑃(𝜆𝜆), 𝑌𝑌~𝑃𝑃𝑃𝑃(𝜇𝜇) and 𝑍𝑍 = 𝑋𝑋 + 𝑌𝑌:

𝑍𝑍~𝑃𝑃𝑃𝑃(𝜆𝜆 + 𝜇𝜇)

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Type I and Type II Errors

Defining Type I and Type II Errors

Type I Error 𝐻𝐻0 is rejected when it is true

Type II Error 𝐻𝐻0 is not rejected when it is false

In hypothesis testing:

significance level = 𝑃𝑃(type I error) = 𝑃𝑃(rejecting H0 |H0 is true)

Power of a Test and Using Type II Errors

In hypothesis testing:

𝑃𝑃(type II error) = 𝑃𝑃(not rejecting H0 | a specific alternative to H0 )

Power = 1 − 𝑃𝑃(type II error)

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Continuous Random Variables

Probability Density Function (PDF)

For continuous random variable 𝑋𝑋 with pdf 𝑓𝑓(𝑥𝑥):

𝑏𝑏
𝑃𝑃(𝑎𝑎 < 𝑥𝑥 < 𝑏𝑏) = � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑
𝑎𝑎

∞
� 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 = 1
−∞

𝑓𝑓(𝑥𝑥) ≥ 0 for all 𝑥𝑥

Median and Quartiles of a Given Probability Density Function

𝑄𝑄1
1
Lower quartile (𝑄𝑄1 ) � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 =
−∞ 4

𝑀𝑀
1
Median � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 =
−∞ 2

𝑄𝑄3
3
Upper quartile (𝑄𝑄3 ) � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 =
−∞ 4

Expectation and Variance of a Continuous Random Variable

∞
𝐸𝐸(𝑋𝑋) = � 𝑥𝑥 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑
−∞

∞
𝐸𝐸(𝑋𝑋 2 ) = � 𝑥𝑥 2 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑
−∞

𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) = 𝐸𝐸(𝑋𝑋 2 ) − (𝐸𝐸(𝑋𝑋))2

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Functions of a Continuous Random Variable

𝐸𝐸(𝑎𝑎𝑎𝑎 + 𝑏𝑏) = 𝑎𝑎𝑎𝑎(𝑋𝑋) + 𝑏𝑏

Linear Transformations
𝑉𝑉𝑉𝑉𝑉𝑉(𝑎𝑎𝑎𝑎 + 𝑏𝑏) = 𝑎𝑎2 𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋)

∞
General Function 𝐸𝐸�𝑔𝑔(𝑋𝑋)� = � 𝑔𝑔(𝑥𝑥)𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑
−∞

Expectation and Variance of the Sum of Two Independent Random Variables

𝐸𝐸(𝑋𝑋 + 𝑌𝑌) = 𝐸𝐸(𝑋𝑋) + 𝐸𝐸(𝑌𝑌)

𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋 + 𝑌𝑌) = 𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) + 𝑉𝑉𝑉𝑉𝑉𝑉(𝑌𝑌)

Cumulative Distribution Function

For a continuous distribution:

𝑥𝑥
𝑃𝑃(𝑋𝑋 ≤ 𝑥𝑥) = 𝐹𝐹(𝑥𝑥) = � 𝑓𝑓(𝑡𝑡) 𝑑𝑑𝑑𝑑
−∞

𝑑𝑑
𝑓𝑓(𝑥𝑥) = 𝐹𝐹(𝑥𝑥)
𝑑𝑑𝑑𝑑

Rectangular Distribution

For the variable 𝑋𝑋 following a

rectangular distribution over [𝑎𝑎, 𝑏𝑏]
1
𝑓𝑓(𝑥𝑥) = 𝑏𝑏−𝑎𝑎 for 𝑎𝑎 < 𝑥𝑥 < 𝑏𝑏

𝑎𝑎 + 𝑏𝑏
𝐸𝐸(𝑋𝑋) =
2

(𝑏𝑏 − 𝑎𝑎)2
𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) =
12

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Chi squared Tests for Association

Chi squared Values and Degrees of Freedom

Expected frequency 𝑟𝑟𝑟𝑟𝑟𝑟 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 × 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡

𝐸𝐸𝑖𝑖 =
𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡
Chi squared value 2
(𝑂𝑂𝑖𝑖 − 𝐸𝐸𝑖𝑖 )2
𝜒𝜒𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 =�
𝐸𝐸𝑖𝑖
𝑖𝑖

Degrees of freedom 𝑣𝑣 = (𝑛𝑛𝑛𝑛. 𝑜𝑜𝑜𝑜 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 − 1)(𝑛𝑛𝑛𝑛. 𝑜𝑜𝑜𝑜 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 − 1)

Condition on Expected Frequency

2
For 𝐸𝐸𝑖𝑖 > 5: 𝜒𝜒𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 ~𝜒𝜒𝑣𝑣2

Yates’ Correction

(|𝑂𝑂𝑖𝑖 − 𝐸𝐸𝑖𝑖 | − 0.5)2

For 𝑣𝑣 = 1: 2
𝜒𝜒Yates =�
𝐸𝐸𝑖𝑖

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Exponential Distribution

Probability Density Function and Cumulative Distribution Function

For 𝑋𝑋~ exp(𝜆𝜆)

𝑓𝑓(𝑥𝑥) = 𝜆𝜆𝑒𝑒 −𝜆𝜆𝜆𝜆 for 𝑥𝑥 > 0

𝐹𝐹(𝑥𝑥) = 1 − 𝑒𝑒 −𝜆𝜆𝜆𝜆

Mean and Variance of an Exponential Distribution

1
Mean 𝐸𝐸(𝑋𝑋) =
𝜆𝜆

1
Variance 𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) =
𝜆𝜆2

Inference – One Sample t-Distribution

Testing for the Mean of a Normal Distribution with Unknown Variance

𝑋𝑋� − 𝜇𝜇
𝑇𝑇 = ~𝑡𝑡𝑛𝑛−1
𝑆𝑆
√𝑛𝑛

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
 Confidence Intervals

Confidence interval for 𝜎𝜎 𝜎𝜎

(𝑥𝑥 − 𝑧𝑧 , 𝑥𝑥 + 𝑧𝑧 )
population mean 𝜇𝜇 √𝑛𝑛 √𝑛𝑛

𝜎𝜎
Width of confidence interval 2𝑧𝑧
√𝑛𝑛

Symmetric Confidence Intervals from Small Samples Using the t-Distribution

For normally distributed sample with a small sample size and

estimated variance 𝑠𝑠 2 , the 𝑐𝑐% CI for population mean:
𝑠𝑠 𝑠𝑠
𝑥𝑥̅ − 𝑡𝑡 < 𝜇𝜇 < 𝑥𝑥̅ + 𝑡𝑡
√𝑛𝑛 √𝑛𝑛
1
𝑐𝑐
where 𝑃𝑃(𝑡𝑡𝑣𝑣 < 𝑡𝑡) = 0.5 + 2
100

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc