Unbiased Estimates (DP IB Applications & Interpretation (AI)) : Revision Note

Dan Finlay

Last updated

Did this video help you?

Unbiased Estimates

What is an unbiased estimator of a population parameter?

  • An estimator is a random variable that is used to estimate a population parameter

    • An estimate is the value produced by the estimator when a sample is used

  • An estimator is called unbiased if its expected value is equal to the population parameter

    • An estimate from an unbiased estimator is called an unbiased estimate

    • This means that the mean of the unbiased estimates will get closer to the population parameter as more samples are taken

  • The sample mean is an unbiased estimate for the population mean

    • x with bar on top equals fraction numerator sum x over denominator n end fraction

  • The sample variance is not an unbiased estimate for the population variance

    • s subscript n superscript 2 equals fraction numerator sum left parenthesis x minus x with bar on top right parenthesis squared over denominator n end fraction equals fraction numerator sum x squared over denominator n end fraction minus open parentheses x with bar on top close parentheses squared 

    • On average the sample variance will underestimate the population variance

    • As the sample size increases the sample variance gets closer to the unbiased estimate

What are the formulae for unbiased estimates of the mean and variance of a population?

  • A sample of n data values (x1, x2, ... etc) can be used to find unbiased estimates for the mean and variance of the population

  • An unbiased estimate for the mean μ of a population can be calculated using

    •  x with bar on top equals fraction numerator sum x over denominator n end fraction

  • An unbiased estimate for the variance σ² of a population can be calculated using

    • s subscript n minus 1 end subscript superscript 2 equals fraction numerator n over denominator n minus 1 end fraction s subscript n superscript 2

    • This is given in the formula booklet

    • This can also be written as s subscript n minus 1 end subscript superscript 2 equals fraction numerator sum left parenthesis x minus x with bar on top right parenthesis squared over denominator n minus 1 end fraction

      • Notice that dividing by n gives a biased estimate but dividing by n minus 1 gives an unbiased estimate

  • Different calculators can use different notations for s subscript n minus 1 end subscript superscript 2

    • sigma subscript n minus 1 end subscript superscript 2, s subscript blank superscript 2s with hat on top subscript blank superscript 2 are notations you might see

    • You may also see the square roots of these

Is sn-1 an unbiased estimate for the standard deviation?

  • Unfortunately sn-1 is not an unbiased estimate for the standard deviation of the population

  • It is better to work with the unbiased variance rather than standard deviation

  • There is not a formula for an unbiased estimate for the standard deviation that works for all populations

    • Therefore you will not be asked to find one in your exam

How do I show the sample mean is an unbiased estimate for the population mean?

  • You do not need to learn this proof

    • It is simply here to help with your understanding

  • Suppose the population of X has mean μ and variance σ²  

  • Take a sample of n observations

    • X1, X2, ..., Xn

    • E(Xi) = μ

  • Using the formula for a linear combination of independent variables:

table attributes columnalign right center left columnspacing 0px end attributes row cell straight E invisible function application open parentheses X with bar on top close parentheses end cell equals cell straight E invisible function application open parentheses fraction numerator X subscript 1 plus X subscript 2 plus blank horizontal ellipsis plus X subscript n over denominator n end fraction close parentheses end cell row blank equals cell fraction numerator straight E invisible function application open parentheses X subscript 1 close parentheses plus straight E invisible function application open parentheses X subscript 2 close parentheses plus blank horizontal ellipsis plus straight E open parentheses X subscript n close parentheses over denominator n end fraction end cell row blank equals cell fraction numerator mu plus mu plus blank horizontal ellipsis plus mu blank over denominator n end fraction end cell row blank equals cell fraction numerator n mu over denominator n end fraction end cell row blank equals mu end table

  • As table attributes columnalign right center left columnspacing 0px end attributes row cell straight E invisible function application open parentheses X with bar on top close parentheses end cell equals mu end table this shows the formula will produce an unbiased estimate for the population mean

Why is there a divisor of n-1 in the unbiased estimate for the variance?

  • You do not need to learn this proof

    • It is simply here to help with your understanding

  • Suppose the population of X has mean μ and variance σ²  

  • Take a sample of n observations

    • X1, X2, ..., Xn

    • E(Xi) = μ

    • Var(Xi) = σ2

  • Using the formula for a linear combination of independent variables:

table attributes columnalign right center left columnspacing 0px end attributes row cell Var invisible function application open parentheses X with bar on top close parentheses end cell equals cell Var invisible function application open parentheses fraction numerator X subscript 1 plus X subscript 2 plus blank horizontal ellipsis plus X subscript n over denominator n end fraction close parentheses end cell row blank equals cell fraction numerator Var invisible function application open parentheses X subscript 1 close parentheses plus Var invisible function application open parentheses X subscript 2 close parentheses plus blank horizontal ellipsis plus Var open parentheses X subscript n close parentheses over denominator n squared end fraction end cell row blank equals cell fraction numerator sigma squared plus sigma squared plus blank horizontal ellipsis plus sigma squared blank over denominator n squared end fraction end cell row blank equals cell fraction numerator n sigma squared over denominator n squared end fraction end cell row blank equals cell sigma squared over n end cell end table

  • It can be shown that straight E left parenthesis X with bar on top squared right parenthesis equals mu squared plus sigma squared over n

    • This comes from rearranging Var invisible function application open parentheses X with bar on top close parentheses equals straight E invisible function application open parentheses X with bar on top squared close parentheses minus open square brackets straight E invisible function application open parentheses X with bar on top close parentheses close square brackets squared

  • It can be shown that straight E left parenthesis X squared right parenthesis equals straight E open parentheses X subscript i squared close parentheses equals mu squared plus sigma squared

    • This comes from rearranging Var invisible function application open parentheses X close parentheses equals straight E invisible function application open parentheses X squared close parentheses minus open square brackets straight E invisible function application open parentheses X close parentheses close square brackets squared

  • Using the formula for a linear combination of independent variables:

table attributes columnalign right center left columnspacing 0px end attributes row cell straight E invisible function application open parentheses S subscript n superscript 2 close parentheses end cell equals cell straight E invisible function application open parentheses fraction numerator sum X subscript i superscript 2 over denominator n end fraction minus X with minus on top squared close parentheses end cell row blank equals cell fraction numerator sum straight E invisible function application left parenthesis X subscript i superscript 2 right parenthesis blank over denominator n end fraction minus straight E invisible function application open parentheses X with bar on top squared close parentheses end cell row blank equals cell fraction numerator sum open parentheses mu squared plus sigma squared close parentheses over denominator n end fraction minus open parentheses mu squared plus sigma squared over n close parentheses end cell row blank equals cell fraction numerator n open parentheses mu squared plus sigma squared close parentheses over denominator n end fraction minus open parentheses mu squared plus sigma squared over n close parentheses end cell row blank equals cell mu squared plus sigma squared minus open parentheses mu squared plus sigma squared over n close parentheses end cell row blank equals cell sigma squared minus sigma squared over n end cell row blank equals cell fraction numerator n sigma squared minus sigma squared over denominator n end fraction end cell row blank equals cell fraction numerator n minus 1 over denominator n end fraction sigma ² end cell end table

  • As straight E open parentheses S subscript n superscript 2 close parentheses not equal to straight sigma squared this shows that the sample variance is not unbiased

    • You need to multiply by fraction numerator n over denominator n minus 1 end fraction

    • straight E open parentheses S subscript n minus 1 end subscript superscript 2 close parentheses equals straight sigma squared

Examiner Tips and Tricks

  • Check the wording of the exam question carefully to determine which of the following you are given:

    • The population variance: sigma squared

    • The sample variances subscript n superscript 2

    • An unbiased estimate for the population variances subscript n minus 1 end subscript superscript 2

Worked Example

The times, X minutes, spent on daily revision of a random sample of 50 IB students from the UK are summarised as follows.

n equals 50

sum x equals 6174

s subscript n superscript 2 equals 1384.3

Calculate unbiased estimates of the population mean and variance of the times spent on daily revision by IB students in the UK.

4-6-2-ib-ai-hl-unbiased-estimates-we-solution
👀 You've read 1 of your 5 free revision notes this week
An illustration of students holding their exam resultsUnlock more revision notes. It's free!

By signing up you agree to our Terms and Privacy Policy.

Already have an account? Log in

Did this page help you?

Dan Finlay

Author: Dan Finlay

Expertise: Maths Lead

Dan graduated from the University of Oxford with a First class degree in mathematics. As well as teaching maths for over 8 years, Dan has marked a range of exams for Edexcel, tutored students and taught A Level Accounting. Dan has a keen interest in statistics and probability and their real-life applications.

Download notes on Unbiased Estimates