What do you think?
Rate this book
Bayesian Statistics the Fun Way: Understanding Statistics and Probability with Star Wars, LEGO, and Rubber Ducks
Fun guide to learning Bayesian statistics and probability through unusual and illustrative examples.
Probability and statistics are increasingly important in a huge range of professions. But many people use data in ways they don't even understand, meaning they aren't getting the most from it. Bayesian Statistics the Fun Way will change that.
This book will give you a complete understanding of Bayesian statistics through simple explanations and un-boring examples. Find out the probability of UFOs landing in your garden, how likely Han Solo is to survive a flight through an asteroid shower, how to win an argument about conspiracy theories, and whether a burglary really was a burglary, to name a few examples.
By using these off-the-beaten-track examples, the author actually makes learning statistics fun. And you'll learn real skills, like how
- How to measure your own level of uncertainty in a conclusion or belief
- Calculate Bayes theorem and understand what it's useful for
- Find the posterior, likelihood, and prior to check the accuracy of your conclusions
- Calculate distributions to see the range of your data
- Compare hypotheses and draw reliable conclusions from them
Next time you find yourself with a sheaf of survey results and no idea what to do with them, turn to Bayesian Statistics the Fun Way to get the most value from your data.
Probability and statistics are increasingly important in a huge range of professions. But many people use data in ways they don't even understand, meaning they aren't getting the most from it. Bayesian Statistics the Fun Way will change that.
This book will give you a complete understanding of Bayesian statistics through simple explanations and un-boring examples. Find out the probability of UFOs landing in your garden, how likely Han Solo is to survive a flight through an asteroid shower, how to win an argument about conspiracy theories, and whether a burglary really was a burglary, to name a few examples.
By using these off-the-beaten-track examples, the author actually makes learning statistics fun. And you'll learn real skills, like how
- How to measure your own level of uncertainty in a conclusion or belief
- Calculate Bayes theorem and understand what it's useful for
- Find the posterior, likelihood, and prior to check the accuracy of your conclusions
- Calculate distributions to see the range of your data
- Compare hypotheses and draw reliable conclusions from them
Next time you find yourself with a sheaf of survey results and no idea what to do with them, turn to Bayesian Statistics the Fun Way to get the most value from your data.
256 pages, Paperback
Published July 9, 2019
560 people are currently reading
1,840 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 55 reviews
October 15, 2019
P(You will enjoy this book | you have no stats background) = .8679
July 28, 2021
This is a great book if you want to find out what all this Bayesian stuff is all about, without getting bogged down in lots of derivations. If you are used to reading mathematical texts you might find it overly simplified, as every step in the journey from A to B is explained in detail, with lots of concrete (and entertaining) examples. Personally, even though I am used to mathematical texts, I found it very soothing to be spoon-fed in delicate little bites for a change. It's clearly and engagingly written with compact chapters, I could read it very quickly, and afterwards I definitely knew a lot more about the Bayesian approach. What's not to like? There are some errors that could have been caught by the proof-readers (incorrectly named variable, referring to colours on a black and white diagram, incorrect summing of a distribution) but nothing that would hinder understanding more than momentarily.
July 24, 2024
Bayesian Statistics is a method you can use to make sense of the world around you using data. Bayesian Statistics uses something called a "Prior" to calculate your degree of belief in an event. An essential process in using Bayesian Statistics is updating your beliefs according to the data you collect. The other paradigm in statistics is Frequentist. You run as many trials as possible to obtain the data you require. However, in some cases, this isn't possible. Take an election as an example. They only happen once, so you can't run multiple elections to discover how typical a result is.
Will Kurt makes it fun by referring to Gacha games, UFOs, and Star Wars movies. Gacha games are popular in Japan, but I believe people play them everywhere now.
The math is easy to follow. The author explains everything in each equation and breaks it down well. The book does have some calculus in it.
At the end of each chapter is a selection of practice problems. You can use them to test your comprehension of the material. For example, chapter two covers classic probability with coin flips and dice rolls. One question it asks is, what is the probability of rolling two 6-sided dies and getting a number greater than 7?
Kurt's intent with the book was to make a crash course on Bayesian Statistics. He wanted to make a book you could read on a plane ride and have the essentials down. Although the book takes examples from popular culture, it uses statistical methods and tools to solve the problems.
I enjoyed the book. Thanks for reading my review, and see you next time.
Will Kurt makes it fun by referring to Gacha games, UFOs, and Star Wars movies. Gacha games are popular in Japan, but I believe people play them everywhere now.
The math is easy to follow. The author explains everything in each equation and breaks it down well. The book does have some calculus in it.
At the end of each chapter is a selection of practice problems. You can use them to test your comprehension of the material. For example, chapter two covers classic probability with coin flips and dice rolls. One question it asks is, what is the probability of rolling two 6-sided dies and getting a number greater than 7?
Kurt's intent with the book was to make a crash course on Bayesian Statistics. He wanted to make a book you could read on a plane ride and have the essentials down. Although the book takes examples from popular culture, it uses statistical methods and tools to solve the problems.
I enjoyed the book. Thanks for reading my review, and see you next time.
January 15, 2022
Will Kurt introduces Bayesian statistics, one of two broad categories of , also known as subjective or evidential probability. (The other category being , also known as objective or aleatoric probability.) As Bayesian reasoning applies to any problem domain in which people update their beliefs as they acqure new evidence, it shows up in many areas of science (and even history, as when Richard Carrier uses it to interpretate historical evidence; see for example his ).
Kurt explains the subject via examples and exercise problems in every chapter. The serious learner will probably want to install the which the book uses heavily. This isn't absolutely necessary for making sense of this book, but at some point in your statistical journey you're probably going to tackle a problem requiring more than simple arithmetic. When that happens, R looks like a good choice, as it's free to use and has immense statistical capabilities.
My only complaint, or perhaps caution, is about this passage in Chapter 18: When Data Doesn’t Convince You:
In this passage, Kurt leaves the orderly confines of mathematics and enters the topsy-turvy world of psychology. He's speculating on the psychology of belief formation and modification. Now, perhaps human brains evolved to operate in some ways like Bayesian learning machines, but one doesn't have to read far in psychology to learn that individuals differ from each other - a lot. When it comes to human psychology there are very few universal traits. Posing the question "What evidence would change your mind?" to different people may not be very informative. People differ in their capacity to imagine what sort of evidence is even possible, let alone how they might respond to it. After all, if someone strongly believes something, they don't believe there is evidence against it. So the question is asking them to imagine something they don't believe can happen - not everyone can do that. It's like asking people from the 1920s to imagine how the Internet might change their lives. Science fiction written a century ago shows how even the more imaginative people struggled to produce an accurate picture of the future they hadn't experienced.
And while Kurt doesn't mention it by name, he seemingly alludes to the , a name for the finding that given evidence against their beliefs, people can reject the evidence and believe even more strongly. According to Wikipedia, early reports of this phenomenon have not stood up to replication:
People change their beliefs all the time, even strongly-held beliefs. For example as I write this review, the COVID-19 pandemic continues to rage two years in. In the wealthier countries, highly safe and effective vaccines against COVID-19 have been abundant for months, free of charge, reducing a person's probability of experiencing the range of undesirable outcomes: infection by SARS-CoV-2 (and thereby becoming able to spread the infection to others); symptomatic illness; severe illness; hospitalization; death; or recovery followed by "long COVID" symptoms. Data from the CDC show a 20-fold reduction in death rates for fully vaccinated and boosted persons compared to unvaccinated persons for October, 2021 when the delta variant was dominant. The newer and more infectious omicron variant is becoming dominant. It seems more able to produce "breakthrough" infections in vaccinated individuals but the vaccines are still reducing the odds of the most serious outcomes (hospitalization and death). The result is that virtually all medical experts continue to urge everyone to get their free vaccination. And of the thousands of people dying from COVID-19 every week, the vast majority are unvaccinated.
Yet millions of people have succumbed to a flood of disinformation on social media and right-wing media outlets casting doubt on the safety and efficacy of vaccines while promoting quack remedies. Many of the resulting vaccine deniers insist they will never get vaccinated. If you've ever had the pleasure of speaking to one, you've probably found they seem impervious to facts and evidence.
But not all evidence is created equal. There may not be anything you can say to a vaccine denier in words that compares to personally experiencing COVID-19. As most of the people now in US hospitals for COVID-19 chose to not get vaccinated, hospitals are full of vaccine deniers who are experiencing a form of evidence stronger than mere words. Some of these deniers stick to their guns to their dying breath, denying that they have COVID at all, sometimes becoming verbally abusive toward the health care workers who are trying to save their lives. Others recognize their mistake, and express regret. Some survivors become vaccine evangelists, urging others not to make the same mistake.
Thus the question "What evidence would change your mind?" may be useless, for example if you pose it to a vaccine denier who cannot, while healthy, imagine what dying of COVID-19 feels like. The true answer to the question might be "I would change my mind if I end up in the ICU with COVID-19 and I'm surrounded by medical professionals telling me I probably wouldn't be there if I had taken my free vaccination," but your interlocutor may be unable to imagine that experience, given only your abstract hypothetical question.
The only reliable way to discover how someone will react to evidence is to share it. And don't worry about the backfire effect - according to the best evidence available, you're probably not going to make someone disbelieve facts harder by giving them the facts.
People change their beliefs when the cost of maintaining a belief exceeds the cost of changing it. "Doubling down" on a previously held belief requires some effort, which means it carries a cost. The more evidence a person must conciously reject, the more they have to work at it. When someone experiences evidence they find overwhelming - such as experiencing a deadly disease directly - they may experience "catastrophic dissonance" and find it impossible to continue denying facts. That's an expensive way to correct errors in reasoning, of course. Thus one strategy might be to find cheaper routes to the same experience, such as to introduce the vaccine denier to a former fellow denier who learned to believe medical science the hard way. As the pandemic continues to rip through denier ranks, this kind of social evidence against the denier position continues to pile up.
Kurt explains the subject via examples and exercise problems in every chapter. The serious learner will probably want to install the which the book uses heavily. This isn't absolutely necessary for making sense of this book, but at some point in your statistical journey you're probably going to tackle a problem requiring more than simple arithmetic. When that happens, R looks like a good choice, as it's free to use and has immense statistical capabilities.
My only complaint, or perhaps caution, is about this passage in Chapter 18: When Data Doesn’t Convince You:
The danger of nonfalsifiable beliefs in Bayesian reasoning isn’t just that they can’t be proved wrong—it’s that they are strengthened even by evidence that seems to contradict them. Rather than persisting in trying to convince you, your friend should have first asked, “What can I show you that would change your mind?� If your reply had been that nothing could change your mind, then your friend would be better off not presenting you with more evidence.
So, the next time you argue with a relative over politics or conspiracy theories, you should ask them: “What evidence would change your mind?� If they have no answer to this, you’re better off not trying to defend your views with more evidence, as it will only increase your relative’s certainty in their belief.
In this passage, Kurt leaves the orderly confines of mathematics and enters the topsy-turvy world of psychology. He's speculating on the psychology of belief formation and modification. Now, perhaps human brains evolved to operate in some ways like Bayesian learning machines, but one doesn't have to read far in psychology to learn that individuals differ from each other - a lot. When it comes to human psychology there are very few universal traits. Posing the question "What evidence would change your mind?" to different people may not be very informative. People differ in their capacity to imagine what sort of evidence is even possible, let alone how they might respond to it. After all, if someone strongly believes something, they don't believe there is evidence against it. So the question is asking them to imagine something they don't believe can happen - not everyone can do that. It's like asking people from the 1920s to imagine how the Internet might change their lives. Science fiction written a century ago shows how even the more imaginative people struggled to produce an accurate picture of the future they hadn't experienced.
And while Kurt doesn't mention it by name, he seemingly alludes to the , a name for the finding that given evidence against their beliefs, people can reject the evidence and believe even more strongly. According to Wikipedia, early reports of this phenomenon have not stood up to replication:
However, subsequent research has since failed to replicate findings supporting the backfire effect. One study conducted out of the Ohio State University and George Washington University studied 10,100 participants with 52 different issues expected to trigger a backfire effect. While the findings did conclude that individuals are reluctant to embrace facts that contradict their already held ideology, no cases of backfire were detected. The backfire effect has since been noted to be a rare phenomenon rather than a common occurrence.
People change their beliefs all the time, even strongly-held beliefs. For example as I write this review, the COVID-19 pandemic continues to rage two years in. In the wealthier countries, highly safe and effective vaccines against COVID-19 have been abundant for months, free of charge, reducing a person's probability of experiencing the range of undesirable outcomes: infection by SARS-CoV-2 (and thereby becoming able to spread the infection to others); symptomatic illness; severe illness; hospitalization; death; or recovery followed by "long COVID" symptoms. Data from the CDC show a 20-fold reduction in death rates for fully vaccinated and boosted persons compared to unvaccinated persons for October, 2021 when the delta variant was dominant. The newer and more infectious omicron variant is becoming dominant. It seems more able to produce "breakthrough" infections in vaccinated individuals but the vaccines are still reducing the odds of the most serious outcomes (hospitalization and death). The result is that virtually all medical experts continue to urge everyone to get their free vaccination. And of the thousands of people dying from COVID-19 every week, the vast majority are unvaccinated.
Yet millions of people have succumbed to a flood of disinformation on social media and right-wing media outlets casting doubt on the safety and efficacy of vaccines while promoting quack remedies. Many of the resulting vaccine deniers insist they will never get vaccinated. If you've ever had the pleasure of speaking to one, you've probably found they seem impervious to facts and evidence.
But not all evidence is created equal. There may not be anything you can say to a vaccine denier in words that compares to personally experiencing COVID-19. As most of the people now in US hospitals for COVID-19 chose to not get vaccinated, hospitals are full of vaccine deniers who are experiencing a form of evidence stronger than mere words. Some of these deniers stick to their guns to their dying breath, denying that they have COVID at all, sometimes becoming verbally abusive toward the health care workers who are trying to save their lives. Others recognize their mistake, and express regret. Some survivors become vaccine evangelists, urging others not to make the same mistake.
Thus the question "What evidence would change your mind?" may be useless, for example if you pose it to a vaccine denier who cannot, while healthy, imagine what dying of COVID-19 feels like. The true answer to the question might be "I would change my mind if I end up in the ICU with COVID-19 and I'm surrounded by medical professionals telling me I probably wouldn't be there if I had taken my free vaccination," but your interlocutor may be unable to imagine that experience, given only your abstract hypothetical question.
The only reliable way to discover how someone will react to evidence is to share it. And don't worry about the backfire effect - according to the best evidence available, you're probably not going to make someone disbelieve facts harder by giving them the facts.
People change their beliefs when the cost of maintaining a belief exceeds the cost of changing it. "Doubling down" on a previously held belief requires some effort, which means it carries a cost. The more evidence a person must conciously reject, the more they have to work at it. When someone experiences evidence they find overwhelming - such as experiencing a deadly disease directly - they may experience "catastrophic dissonance" and find it impossible to continue denying facts. That's an expensive way to correct errors in reasoning, of course. Thus one strategy might be to find cheaper routes to the same experience, such as to introduce the vaccine denier to a former fellow denier who learned to believe medical science the hard way. As the pandemic continues to rip through denier ranks, this kind of social evidence against the denier position continues to pile up.
February 16, 2025
This is a good and (fairly) gentle introduction to an interesting field. It's not quite as gentle as the cover suggests, but it doesn't rush into anything like textbooks written for university classes often do. It slowly explains every step, with interesting descriptions. You would benefit from some prior knowledge of mathematical notation when reading. It does explain its integrals and other symbols as it goes, but it's probably a lot to take in if you're unfamiliar with post-GCSE mathematics.
Also, the Star, Wars, LEGO and Rubber Ducks thing in the sub-heading is not really representative of the tone of the book. This isn't as light-hearted as the O'Reilly Head First books for example. They are usually great guides, but their style can get tiresome. Will Kurt has taken a middle ground, with a style that is light without being excessively so.
Also, the Star, Wars, LEGO and Rubber Ducks thing in the sub-heading is not really representative of the tone of the book. This isn't as light-hearted as the O'Reilly Head First books for example. They are usually great guides, but their style can get tiresome. Will Kurt has taken a middle ground, with a style that is light without being excessively so.
May 10, 2023
This book describes introductory topics in Bayesian statistics in a brisk and accessible manner. The exposition starts with basic probability and then gets into conditional probability and Bayes theorem. It then talks about parameter estimation (the exposition about mean as an estimation is nice) and hypothesis testing. Through out, it keeps the contents both accessible and intuitive, which I was happy with when compared to other books I have tried to read on this subject. Of course, this means it sacrifices mathematical rigor, which it explicitly does not promise in the title :) This also means that reader will have to read more elsewhere to get into Bayesian data analysis or deeper+applied aspects of Bayesian statistics.
Overall, a good introductory book on probability and Bayesian statistics.
Overall, a good introductory book on probability and Bayesian statistics.
September 19, 2019
A really interesting approach to teaching bayesian statistics. My biggest complaint is that it does not stick with "degree of belief" interpretation of probability consistently, often appealing to frequentist ideas.
March 13, 2021
A really great, well-written book, pitched at the perfect level for me (a human who has forgotten all the calculus they struggled through a decade or two ago). Each chapter walks through a complete, well-defined step, with really great, illuminating examples. The only problems I had were keeping all the vocab straight (I immediately forget exactly what the "prior" is every single time I look it up), and failing to grok the actual Bayes' Theorem. I felt like Moses seeing the promised land--I just couldn't get there, even though Kurt spends two chapters on it.
October 4, 2024
Clearly written by someone who has spent too much time on "that" forum.
A great book that serves as a revision of basic statistics (probabilities, inference, beta, binomial and normal distributions, Bayesian reasoning, and hypothesis testing using Bayesian reasoning.) but from a Bayesian instead of a Frequentist lens.
A great book that serves as a revision of basic statistics (probabilities, inference, beta, binomial and normal distributions, Bayesian reasoning, and hypothesis testing using Bayesian reasoning.) but from a Bayesian instead of a Frequentist lens.
January 30, 2022
Well organization particularly chapters 4 to 6 and 16 to 18 well written!
June 14, 2020
Excellent intro!
May 3, 2023
Really great read! Intuitive and clear concepts for understanding and implementing Bayesian reasoning in everyday life.
April 22, 2021
Nice and informal intro to Bayesian thinking.
February 13, 2021
I’ve been wanting to learn more about Bayesian statistics since grad school but I never got around to it. I happened to walk past this book at a local bookstore and it caught my eye. It is not too long; just the right length for a useful introduction. It was perfect for finally scratching that Bayesian itch! :)
March 3, 2020
This is a wonderful book that hold your hand and guide you through Bayesian Statistics from first principle. I wish I had read this book first before diving and wasting so many time on others books, which leave me nothing but a sour taste for Statistics.
While this book is certainly not an extensive textbook cover all topics of statistics, it will give you enough motivation and, more importantly, love to get you down the Statistics rabbit's hole.
I love how each concept is carefully aided by a well-designed real life, relevant example, rather than just abstract, vague description like from other books.
While this book is certainly not an extensive textbook cover all topics of statistics, it will give you enough motivation and, more importantly, love to get you down the Statistics rabbit's hole.
I love how each concept is carefully aided by a well-designed real life, relevant example, rather than just abstract, vague description like from other books.
December 19, 2020
This is the kind of book you would hand to someone to bring statistics down to earth for them, making it easy without a lot of math. While people who are already familiar with Bayes or statistics might not get much out of this book, it is perfect for someone who knows nothing about them. While some of the topics I didn’t consider personally fun, I would say that this author did a great job making statistics less scary and less boring. I think this would be perfect for an introduction to statistics class or just to read in your own time if you want to increase your knowledge of Bayes algorithms. If you’ll enjoy that then jump on into “Bayesian Statistics the Fun Way�
Ok, are they gone? I think I got rid of all the people who will only look at the highlight sentences. So now I’m going to sit down and go through all the actual pieces of this book and explain my system. I rated this book on 3 main subjects:
Readability: How easy is this book to read and how quickly could a person finish the whole book
Accuracy: how accurate is the book? How many errors does it make in the code or text?
Subject: Does this book do what the cover or back says it does? Is what the book wants to teach you good?
Bonus Points: This is just for outside resources that the book points to directly. I might also add some places where you can find more education from the author such as informative twitter threads or medium articles that the author offers for free.
Readability 5/10:
This book can be fun at times and it does go a long way to make statistics and math something that the average person can get into. The problem with this book is the subject will only allow it to take this so far. You will probably find yourself getting through the first few chapters easily. Taking the book and reading it pieces at a time will actually be a nice and relaxing way to educate yourself on statistics.
The problem is that you will eventually get to more complicated sections and, if you try to read this quickly or in a single sitting, you will find finishing this book difficult. I read this book over 3 weeks at about 10-15 pages per day. This book is best enjoyed in bites or nibbles rather than a full gulp of a read. I think this is a positive for the book itself and allows any reader who might not fully grasp the material time to digest and ask questions.
Accuracy 10/10:
What can I say other than this book hits it’s mark? This is a technical book for the average person and because of that I couldn’t find anything that really stuck out as debatable or incorrect.
Subject 10/10:
This is the kind of book that someone wrote because they didn’t want the next person to get the same dry education that they got back in college. While this doesn’t mean that the subject is fun in it’s own right, the author does a great job to make it far more palatable than this subject could ever be in most classroom text books. If you’re going to be teaching Bayes in a computer science course or in a statistics class before you move onto something like Kmeans, then this would be the book to teach with. Even first year students without heavy mathematical knowledge will be able to pick it up and stroll through on their own. The funny thing about this book is, while reading the first few chapters, the only thing I could think was “There are some conspiracy theorists out there who could improve their lives and remove their theories with this book.� While I’m hesitant to say “This is a book for conspiracy theorists!� I would say that it will do something to improve your reasoning on the world around you. So if you know a friend who is drifting to close to the crazy then see what throwing this book at them will do.
Bonus points:
+1 To the sections at the back teaching R and Calculus:
I would like to give a shout out to the author who is willing to make a book that the average person can understand, but still put in a section to “Know More.� This is just a tip of the hat as I have seen a lot of authors come into these subjects with the idea of “You don’t need this for my book so I won’t even give you a hint about it. You’ll get that in someone else’s book.� The section is appreciated
About the Author:
After researching the author a little bit, it seems that he has a website where he blogs about data science topics and his next book. He has points where you can get involved and ask him questions directly. So if you enjoyed this book and want to step deeper in then I would recommend googling Will Kurt and hopping into what you find. Who knows you might ask a question that will change his next book.
Ok, are they gone? I think I got rid of all the people who will only look at the highlight sentences. So now I’m going to sit down and go through all the actual pieces of this book and explain my system. I rated this book on 3 main subjects:
Readability: How easy is this book to read and how quickly could a person finish the whole book
Accuracy: how accurate is the book? How many errors does it make in the code or text?
Subject: Does this book do what the cover or back says it does? Is what the book wants to teach you good?
Bonus Points: This is just for outside resources that the book points to directly. I might also add some places where you can find more education from the author such as informative twitter threads or medium articles that the author offers for free.
Readability 5/10:
This book can be fun at times and it does go a long way to make statistics and math something that the average person can get into. The problem with this book is the subject will only allow it to take this so far. You will probably find yourself getting through the first few chapters easily. Taking the book and reading it pieces at a time will actually be a nice and relaxing way to educate yourself on statistics.
The problem is that you will eventually get to more complicated sections and, if you try to read this quickly or in a single sitting, you will find finishing this book difficult. I read this book over 3 weeks at about 10-15 pages per day. This book is best enjoyed in bites or nibbles rather than a full gulp of a read. I think this is a positive for the book itself and allows any reader who might not fully grasp the material time to digest and ask questions.
Accuracy 10/10:
What can I say other than this book hits it’s mark? This is a technical book for the average person and because of that I couldn’t find anything that really stuck out as debatable or incorrect.
Subject 10/10:
This is the kind of book that someone wrote because they didn’t want the next person to get the same dry education that they got back in college. While this doesn’t mean that the subject is fun in it’s own right, the author does a great job to make it far more palatable than this subject could ever be in most classroom text books. If you’re going to be teaching Bayes in a computer science course or in a statistics class before you move onto something like Kmeans, then this would be the book to teach with. Even first year students without heavy mathematical knowledge will be able to pick it up and stroll through on their own. The funny thing about this book is, while reading the first few chapters, the only thing I could think was “There are some conspiracy theorists out there who could improve their lives and remove their theories with this book.� While I’m hesitant to say “This is a book for conspiracy theorists!� I would say that it will do something to improve your reasoning on the world around you. So if you know a friend who is drifting to close to the crazy then see what throwing this book at them will do.
Bonus points:
+1 To the sections at the back teaching R and Calculus:
I would like to give a shout out to the author who is willing to make a book that the average person can understand, but still put in a section to “Know More.� This is just a tip of the hat as I have seen a lot of authors come into these subjects with the idea of “You don’t need this for my book so I won’t even give you a hint about it. You’ll get that in someone else’s book.� The section is appreciated
About the Author:
After researching the author a little bit, it seems that he has a website where he blogs about data science topics and his next book. He has points where you can get involved and ask him questions directly. So if you enjoyed this book and want to step deeper in then I would recommend googling Will Kurt and hopping into what you find. Who knows you might ask a question that will change his next book.
September 5, 2022
Il libro non si limita a spiegare il teorema di Bayes. Prova a trattare anche argomenti connessi, come la comparazione di ipotesi e l’A/B test. Purtroppo, è difficile riuscirci compattando ogni capitolo in 7/8 pagine.
La prima parte è ben fatta. Vengono introdotti i concetti base della probabilità (come P(A,B) e P(A OR B)), si spiega la differenza tra probabilità (dove si parte dalle probabilità per sapere esiti eventi) e statistica (dove si parte dai dati per calcolare probabilità ). Si tratta della distribuzione di probabilità polinomiale (ad esempio per calcolare la probabilità che due monete su cinque siano testa), della distribuzione beta (per rappresentare infinite ipotesi) e delle probabilità condizionali (probabilità degli eventi dipendono tra loro).
L’autore affronta il teorema da diverse prospettive. Dato quello che osservo (D), l’esperienza pregressa (X), e la nuova ipotesi (H), si può utilizzare il teorema per calcolare:
- Quanto le mie credenze spiegano quello che osservo? P(D|H,X)
- Quanto quello che osservo supporta le mie credenze? P(H,X|D)
Ho trovato utile l’utilizzo dei lego per rappresentare il concetto di probabilità condizionata.
In questo contesto si nota l'importanza del calcolo differenziale, generalmente associato ad una variabile che cambia nel tempo (ad esempio la posizione), qui viene usata per calcolare la probabilità che un valore sia inferiore ad una soglia. Significa cioè calcolare l'integrale della funziona di probabilità in quel punto (in fondo al libro c'è un'appendice che introduce le nozioni base del calcolo differenziale.)
Purtroppo, la seconda parte del libro si fa troppo complessa e condensata per venir compresa del tutto.
La prima parte è ben fatta. Vengono introdotti i concetti base della probabilità (come P(A,B) e P(A OR B)), si spiega la differenza tra probabilità (dove si parte dalle probabilità per sapere esiti eventi) e statistica (dove si parte dai dati per calcolare probabilità ). Si tratta della distribuzione di probabilità polinomiale (ad esempio per calcolare la probabilità che due monete su cinque siano testa), della distribuzione beta (per rappresentare infinite ipotesi) e delle probabilità condizionali (probabilità degli eventi dipendono tra loro).
L’autore affronta il teorema da diverse prospettive. Dato quello che osservo (D), l’esperienza pregressa (X), e la nuova ipotesi (H), si può utilizzare il teorema per calcolare:
- Quanto le mie credenze spiegano quello che osservo? P(D|H,X)
- Quanto quello che osservo supporta le mie credenze? P(H,X|D)
Ho trovato utile l’utilizzo dei lego per rappresentare il concetto di probabilità condizionata.
In questo contesto si nota l'importanza del calcolo differenziale, generalmente associato ad una variabile che cambia nel tempo (ad esempio la posizione), qui viene usata per calcolare la probabilità che un valore sia inferiore ad una soglia. Significa cioè calcolare l'integrale della funziona di probabilità in quel punto (in fondo al libro c'è un'appendice che introduce le nozioni base del calcolo differenziale.)
Purtroppo, la seconda parte del libro si fa troppo complessa e condensata per venir compresa del tutto.
November 4, 2021
The concepts in this book are very important, especially if you want to change from being convinced about your beliefs to learning how to test them. Unfortunately, although the examples the author uses to explain concepts are fun the analysis and statistical concepts are definitely not easy to learn.
If you accept your ignorance of math, computer programming and statistical analysis you'll enjoy the book more. For example learn about how to compare hypotheses using a famous episode from The Twilight Zone. Learn about probability with Legos. Learn, using statistical analysis, why trying to reason with someone who is certain of their position only hardens their resolve.
That last one alone may save you this coming Thanksgiving!
I rate this three stars (instead of higher) mainly because the math, statistics and computer knowledge remains somewhat daunting. Moreover, he sticks to Bayesian terminology (prior odds, posterior odds, etc.) that I never could easily keep straight. I would have preferred he try to use plain English.
Glad I read it, but I recommend it with caution.
If you accept your ignorance of math, computer programming and statistical analysis you'll enjoy the book more. For example learn about how to compare hypotheses using a famous episode from The Twilight Zone. Learn about probability with Legos. Learn, using statistical analysis, why trying to reason with someone who is certain of their position only hardens their resolve.
That last one alone may save you this coming Thanksgiving!
I rate this three stars (instead of higher) mainly because the math, statistics and computer knowledge remains somewhat daunting. Moreover, he sticks to Bayesian terminology (prior odds, posterior odds, etc.) that I never could easily keep straight. I would have preferred he try to use plain English.
Glad I read it, but I recommend it with caution.
May 22, 2023
While this may be the college equivalent of a children's book, I want to remind anyone reading this review that children's books are nothing to be scoffed at. Due to my inability to fully grasp the regular textbooks of my current schooling program, I had to resort to this silly book on statistics in preparation for my Bayesian Econometrics course. I am still absolutely petrified by the 'econometric' portion of that course title, but I am no longer afraid of the 'Bayesian' portion.
This book breaks down statistics into *very* short chapters that can be read in less than 15 minutes each. Step by step, the author introduces not only the basics of Bayesian stats, but also frequentist statistics as well. I feel this book has given me a solid foundation in the statistical theory necessary to keep my head above water in my upcoming course. This book is worth it for anyone who has a weak foundation in statistics and needs a few spoonfuls of sugar to get that medicine down.
This book breaks down statistics into *very* short chapters that can be read in less than 15 minutes each. Step by step, the author introduces not only the basics of Bayesian stats, but also frequentist statistics as well. I feel this book has given me a solid foundation in the statistical theory necessary to keep my head above water in my upcoming course. This book is worth it for anyone who has a weak foundation in statistics and needs a few spoonfuls of sugar to get that medicine down.
November 8, 2021
Attempts to simplify statistics, but too dumb to work. I only got to page 27, because 26-27 was so bad. Trying to show joint probability, the author explained how you can figure out if you'll be late with the two options of traveling by bus or train for a meeting.
Let's start with the ignorance of somebody scheduling to travel in a way that provides no slack time, and just deal with the statistics of figuring out if you're late. "Since you be late only if both the bus and train are late." The author thinks this makes sense? This is Schrodinger's Journey, where you can be both on the bus and on the train at the same time? No! You're late only if the method you choose is late, so you pick the method that's late less often.
The problem setups before that weren't great, but that was the final straw. Avoid this book.
Let's start with the ignorance of somebody scheduling to travel in a way that provides no slack time, and just deal with the statistics of figuring out if you're late. "Since you be late only if both the bus and train are late." The author thinks this makes sense? This is Schrodinger's Journey, where you can be both on the bus and on the train at the same time? No! You're late only if the method you choose is late, so you pick the method that's late less often.
The problem setups before that weren't great, but that was the final straw. Avoid this book.
January 10, 2020
I really liked this book. Kurt is a good technical writer who can clearly explain boring concepts in the most concise ways. I also liked the way the book moves from topics to topics. There's correlation and connection that builds upon knowledge gained from one chapter to the next. I made a note to re-read chapter 11 where Kurt explained the normal distribution concept with the example of a villain who has 6 fuses to set off a bomb. And how we can use statistics to determine the strength of our belief in the uncertainties.
And I liked the simple R codes he scattered for us to use. When I got the same results in R, I was elated. Best 5 seconds no money can buy!
And I liked the simple R codes he scattered for us to use. When I got the same results in R, I was elated. Best 5 seconds no money can buy!
May 11, 2021
The book provides a solid foundation for understanding Bayesian statistics.
For example, if there is a new topic to discuss the first chapter will explain it using mathematical terms and the second chapter will be used to explain the topic using scenarios the everyday person might be familiar with.
But the essential premise of this book for me is that since we all have priors (existing beliefs about certain situations), we might be inclined to give out different perspectives for some scenarios however it should be noted that if we are presented with more valid data that may or may not disprove or strengthen our priors, we should be able to change our priors accordingly given to the data presented.
The most important lesson I learned is that our opinions should be falsifiable and we as human beings should be able to adapt.
For example, if there is a new topic to discuss the first chapter will explain it using mathematical terms and the second chapter will be used to explain the topic using scenarios the everyday person might be familiar with.
But the essential premise of this book for me is that since we all have priors (existing beliefs about certain situations), we might be inclined to give out different perspectives for some scenarios however it should be noted that if we are presented with more valid data that may or may not disprove or strengthen our priors, we should be able to change our priors accordingly given to the data presented.
The most important lesson I learned is that our opinions should be falsifiable and we as human beings should be able to adapt.
This entire review has been hidden because of spoilers.
March 23, 2023
That book is not a breakthrough in deriving mathematics or teaching you how to code with PyMc Libraries, but rather a great help for explaining why Bayesian Approaches for Analysis make sense in a business environment.
Great recommendation develop understanding and context knowledge, rather than focusing on the tools too much.
Best read as an introduction or if prior knowledge is already available to get a new perpective.
Great recommendation develop understanding and context knowledge, rather than focusing on the tools too much.
Best read as an introduction or if prior knowledge is already available to get a new perpective.
October 30, 2020
A light and enjoyable introduction to Bayesian statistics. I liked that most examples could be understood via the binomial distribution and a beta prior. However, it remains (by choice) a bit on the surface.
Caution: there are some slight errors in the formulas and terminology. Not to the extent that the computations would be wrong, but a mathematician might be bothered!
Caution: there are some slight errors in the formulas and terminology. Not to the extent that the computations would be wrong, but a mathematician might be bothered!
January 20, 2024
This book is a popular science read, geared towards beginners. If you want to learn more in-depth knowledge of statistics or algorithms, this book is not a good choice. However, it does provide a Bayesian perspective to explain many interesting decision-making scenarios in life. Additionally, it offers an explanation as to why some person cannot be persuaded by data.
August 31, 2020
As a person with statistical training I found this book an easy read but also very informative. The author has his own particular style of teaching and choice of scenarios which might not appeal to everyone but there is no doubt the underlying statistics are sound.
April 3, 2021
If you're interested in stats but overwhelmed by math notations and like more of real-life analogical thinking, this book will be a life-saver!!
The author uses fun real life scenarios to instill the importance of statistical reasoning and how to use it.
The author uses fun real life scenarios to instill the importance of statistical reasoning and how to use it.
March 6, 2022
A Tutorial Work for Bayesian Statistics
This work also has a chapter on Calculus
While Ago, I read another work, to help me visualize
Review: Introducing Statistics
Deus Vult,
Gottfried
This work also has a chapter on Calculus
While Ago, I read another work, to help me visualize
Review: Introducing Statistics
Deus Vult,
Gottfried
December 22, 2022
One of the best hands-on stats/maths books I've ever seen. A joy to read, gives you an intuition to what's going on, lots of examples, and exercises with answers. I might buy a physical copy I loved it so much.
April 3, 2023
A handy book written in plain language.
The author reviews concepts for you to remember in each chapter.
You dont have to do the math notation, if you're just looking for a basic understanding. At the same time, there are exercises for each chapter if you really want to dig in.
The author reviews concepts for you to remember in each chapter.
You dont have to do the math notation, if you're just looking for a basic understanding. At the same time, there are exercises for each chapter if you really want to dig in.
Displaying 1 - 30 of 55 reviews
Can't find what you're looking for?
Get help and learn more about the design.