- Forums
###### Negative Binomial Question

watch this thread

1 week ago

There are twelve green and eight yellow balls in a bag. One ball is drawn from the bag at random and

replaced. This is repeated until a yellow ball is drawn. The number of balls drawn in total is X.

(a) Find the mean and variance of X.

(b) Find the probability that X is at most one standard deviation from E(X)

I have found the mean as 5/2 and the variance as 15/4.

I’m not sure how to do part b. I know that with a normal distribution it would be 68 percent but I’m not sure how to do it with a negative binomial. Any help would appreciated. Thanks

replaced. This is repeated until a yellow ball is drawn. The number of balls drawn in total is X.

(a) Find the mean and variance of X.

(b) Find the probability that X is at most one standard deviation from E(X)

I have found the mean as 5/2 and the variance as 15/4.

I’m not sure how to do part b. I know that with a normal distribution it would be 68 percent but I’m not sure how to do it with a negative binomial. Any help would appreciated. Thanks

Original post by Cyanforest

There are twelve green and eight yellow balls in a bag. One ball is drawn from the bag at random and

replaced. This is repeated until a yellow ball is drawn. The number of balls drawn in total is X.

(a) Find the mean and variance of X.

(b) Find the probability that X is at most one standard deviation from E(X)

I have found the mean as 5/2 and the variance as 15/4.

I’m not sure how to do part b. I know that with a normal distribution it would be 68 percent but I’m not sure how to do it with a negative binomial. Any help would appreciated. Thanks

replaced. This is repeated until a yellow ball is drawn. The number of balls drawn in total is X.

(a) Find the mean and variance of X.

(b) Find the probability that X is at most one standard deviation from E(X)

I have found the mean as 5/2 and the variance as 15/4.

I’m not sure how to do part b. I know that with a normal distribution it would be 68 percent but I’m not sure how to do it with a negative binomial. Any help would appreciated. Thanks

Its really a geometric distribution (just one win), but you have the mean and standard deviation so you can look at the probabilities which correspond to values

[mean - std dev, mean + std dev]

There are only ~4 values so you could compute the probabilities and sum them up or use the cumulative geometric

https://en.wikipedia.org/wiki/Geometric_distribution

(edited 1 week ago)

Reply 2

1 week ago

Original post by mqb2766

Its really a geometric distribution (just one win), but you have the mean and standard deviation so you can look at the probabilities which correspond to values

[mean - std dev, mean + std dev]

There are only ~4 values so you could compute the probabilities and sum them up or use the cumulative geometric

[url=https://en.wikipedia.org/wiki/Geometric_distribution]https://en.wikipedia.org/wiki/Geometric_distribution]https://en.wikipedia.org/wiki/Geometric_distribution

[mean - std dev, mean + std dev]

There are only ~4 values so you could compute the probabilities and sum them up or use the cumulative geometric

[url=https://en.wikipedia.org/wiki/Geometric_distribution]https://en.wikipedia.org/wiki/Geometric_distribution]https://en.wikipedia.org/wiki/Geometric_distribution

I have tried finding the probability that x is between 8 and 12 inclusive which gave me 0.0258 (3sf) I’m not sure what to do next to relate this to the standard deviation from the mean? (The answer is 0.8704)

Original post by Cyanforest

I have tried finding the probability that x is between 8 and 12 inclusive which gave me 0.0258 (3sf) I’m not sure what to do next to relate this to the standard deviation from the mean? (The answer is 0.8704)

As thats the main body of the distribution (mean+/-sigma), 0.02 is way too small. What did you actually do? The mean is 2.5 and the stddev ~1.9 so not sure where 8 and 12 comes from?

Edit - also note that the ans comes out of the cumulative directly, without having to sum the individual probabilities.

(edited 1 week ago)

Reply 4

1 week ago

Original post by mqb2766

As thats the main body of the distribution (mean+/-sigma), 0.02 is way too small. What did you actually do? The mean is 2.5 and the stddev ~1.9 so not sure where 8 and 12 comes from?

Ah okay I took the 12 and 8 from the amount of green and yellow balls but that makes no sense. I think I’m confused about what I’m actually meant to do regarding the mean and standard deviation and using the cumulative geometric distribution. 2.5 +/- 1.9 = 4.4/0.6 and I’m not sure how that relates to the probability.

Original post by Cyanforest

Ah okay I took the 12 and 8 from the amount of green and yellow balls but that makes no sense. I think I’m confused about what I’m actually meant to do regarding the mean and standard deviation and using the cumulative geometric distribution. 2.5 +/- 1.9 = 4.4/0.6 and I’m not sure how that relates to the probability.

Yup, thats simply wrong.

E(X) is the mean of the distribution so roughly the "centre". The std dev is the "width" (or half the width of our interval). So were interested in

X in [mean - std dev, mean + std dev]

obv X is discrete but that the interval is roughly [0.5, 4.5] so we want the discrete values in that and either compute the relevant probs and sum them, or use the relevant cumulative.

The mean and std dev give you an idea about where the main body of the distribution is and its width, irrespective of the distribution. Thats a key thing to understand. Tbh, Id get in the habit of sketching the distribution (even if the question doesnt ask for it) so that you force yourself to be clear about what youre actually analysing.

(edited 1 week ago)

Reply 6

1 week ago

Original post by mqb2766

Yup, thats simply wrong.

E(X) is the mean of the distribution so roughly the "centre". The std dev is the "width" (or half the width of our interval). So were interested in

X in [mean - std dev, mean + std dev]

obv X is discrete but that the interval is roughly [0.5, 4.5] so we want the discrete values in that and either compute the relevant probs and sum them, or use the relevant cumulative.

The mean and std dev give you an idea about where the main body of the distribution is and its width, irrespective of the distribution. Thats a key thing to understand. Tbh, Id get in the habit of sketching the distribution (even if the question doesnt ask for it) so that you force yourself to be clear about what youre actually analysing.

E(X) is the mean of the distribution so roughly the "centre". The std dev is the "width" (or half the width of our interval). So were interested in

X in [mean - std dev, mean + std dev]

obv X is discrete but that the interval is roughly [0.5, 4.5] so we want the discrete values in that and either compute the relevant probs and sum them, or use the relevant cumulative.

The mean and std dev give you an idea about where the main body of the distribution is and its width, irrespective of the distribution. Thats a key thing to understand. Tbh, Id get in the habit of sketching the distribution (even if the question doesnt ask for it) so that you force yourself to be clear about what youre actually analysing.

Thank you I understand it now

- Negative Binomial Question
- Edexcel Further Stats 1 Paper 3B 2023
- A level maths
- Why can't you use the negative binomial here?
- Probability Generating Functions
- Further Statistics Edexcel 9FM/3B 16th June
- cumalative negative binomial
- Probability Question
- UCAS Clearing 2023 FAQ and Help Megathread
- Got a Ucas or results related question? Ask a Clearing and Applications Advisor here
- Edexcel A Level Further Mathematics Paper 3B: Further Stats 1 9FM0 3B - 17 Jun 2022
- Maximum likelihood estimator variance
- Why is this binomial expansion valid for all ranges of x?
- A level maths
- Anyone know where to find the 2023 AS Edexcel Further Maths papers?
- Maths help
- OCR A Level Mathematics MEI Pure maths & comprehension H640/03 - 21 Jun 2022 [Chat]
- Learning alevel further maths in 5 months
- OCR A-level Mathematics B (MEI) Paper 2 (H640/02) - 13th June 2023 [Exam Chat]
- A level stats challenge question - help needed

- GCSE Mathematics Study Group 2023-2024
- Step support programme assignment 7 help pls
- Senior Maths Challenge 2023
- A-level Mathematics Study Group 2023-2024
- Do we have to memorise all physics equations for GCSE 2024 ?
- further maths gcse
- Girls Maths Olympiad 2023
- Electrical Engineering maths help with Desmos graphing calculator
- A Level Maths Moments question
- Discrete maths questions
- Plotting data on a graph.
- Hypothesis Testing
- Problem solving the smc 2023
- Maths trig
- Engineering maths help
- Send me your maths questions!
- Maths nerds!
- A Level Mechanics.
- Help With A level Trig
- which calculator; the fx-991cw or fx-991ex

- Physician Associate interview 2024
- Big 4 apprenticeships?!?
- What action should i do
- early bird catches the worm: A thread to force myself up and teach myself discipline!
- Ubc admission requirements
- Korean/Japanese Studies 2024 Applicants
- Picking courses
- Is my uni application strong and what do I need to work on?
- University of Oxford 2024 Undergraduate Applicants Official Thread
- LNAT Section A advice

- Sevenoaks vs Godolphin and Latymer
- Slightly nicer party holidays
- Odd personal statement formatting on UCAS site.
- Official: Hull York Medical School A100 2024 Entry Applicants
- ucas
- Breakfast items hurt/heal game!
- The Cambridge College Hurt/Heal Game [part 2]
- British food hurt/heal game!
- The G20 Leaders Hurt/Heal Game
- The Fast Food Restaurants Hurt/Heal game

- The Official Cambridge Applicants for 2024 Entry Thread
- DWP EO Work Coach
- Deloitte Graduate Scheme 2024
- Official London School of Economics and Political Science 2024 Applicant Thread
- BMAT 2024 Entry Discussions Megathread
- Last Person To Post Here Wins (Part 54)
- Official University of Edinburgh Applicant Thread for 2024
- Boys if you have any questions for us girls we will answer
- Official Imperial College London Offer Holders Thread for 2024 entry
- Girls Vs. Boys (Part 45)

- Medicine and biochem
- Procter & gamble
- 2024 Chase Digital Development Programme
- The Forum Games Hurt/Heal Game
- AWE Graduate Scheme 2022
- How to get an overall grade of 2:2 for uni
- KPMG Graduate Applications 2024
- The honorable non-russell group hurt/heal game
- Exam Board Hurt or Heal Game
- Oxford College hurt/heal

The Student Room and The Uni Guide are both part of The Student Room Group.

© Copyright The Student Room 2023 all rights reserved

The Student Room and The Uni Guide are trading names of The Student Room Group Ltd.

Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: Imperial House, 2nd Floor, 40-42 Queens Road, Brighton, East Sussex, BN1 3XB