Chi-squared Goodness of Fit Test! Extensive video!

Поділитися
Вставка
  • Опубліковано 26 гру 2024

КОМЕНТАРІ • 125

  • @gunjansharma6406
    @gunjansharma6406 4 роки тому +17

    You are the best stats tutor I have ever found in you tube... Love and good wishes from India.. ☺️

  • @amywilderson85
    @amywilderson85 3 роки тому +7

    You solved the mystery of the relationship of chi-0squared and normal distribution for me

  • @pedrocolangelo5844
    @pedrocolangelo5844 3 роки тому +4

    You certainly have one of the best channels on UA-cam, sir. Thank you for sharing your knowledge with us!

  • @elcapitano1633
    @elcapitano1633 5 років тому +84

    You are like the coolest teacher ever! How can we repay your favors? Have you thought about a setting up a patreon page? Your videos are insanely good.

    • @teachingfunn8223
      @teachingfunn8223 4 роки тому

      Yes he is cool

    • @korman9872
      @korman9872 3 роки тому

      Yap he is good when explain

    • @jc_777
      @jc_777 Рік тому +2

      I think my university's Statistics dept should all be fired and all the fund there's left should be transfered to this guy's bank account

    • @alienmathology6251
      @alienmathology6251 2 місяці тому

      @@jc_777 thats crazy

  • @sankimalu
    @sankimalu 4 роки тому +25

    I’m a budding actuary and this was on one of my exams! It is so fulfilling to finally understand why we actually use this test and not just having to memorize a formula to pass an exam...

    • @PunmasterSTP
      @PunmasterSTP 7 місяців тому +1

      How are things going as an actuary?

  • @desnamjp
    @desnamjp 2 роки тому +3

    I am taking my doctoral comprehensive exams next month. I was resigned to failing STATS but your videos are truly giving me confidence. To say Thank you would be grossly understated.

  • @sreeharisureshbabu268
    @sreeharisureshbabu268 Місяць тому +1

    Absolutely phenomenal!

  • @fatriantobong2097
    @fatriantobong2097 3 роки тому

    love how you all the sudden remind us about binomial and chi square, it makes all the things i learned converges to one

  • @ohnotoyota4692
    @ohnotoyota4692 3 роки тому +1

    Excellent, specially the explanation on why we use the formula. Thank you!!!

  • @ehsangoodarzi7341
    @ehsangoodarzi7341 4 роки тому +1

    the best explanation of this subject I've ever seen on the net. pretty outstanding. i appreciate that. you put an end on my vague understanding of chi-squared distribution.

  • @justsomegirlwithoutamustac5837
    @justsomegirlwithoutamustac5837 2 роки тому

    UR VIDEOS R SO FREAKING AWESOME PLZ KEEP MAKING CONTENT LIKE THIS

  • @batatambor
    @batatambor 4 роки тому +1

    Excellent! I've been searching the normal explanation for a long time and it's really comforting to finally understand why the goodness of fit test works.

  • @joserobertopacheco298
    @joserobertopacheco298 3 роки тому +2

    Writing from Brazil, I would like to say that this is one of the best mathematical videos I have ever watched. Thank you and congratulations.

  • @johnsteedman7937
    @johnsteedman7937 Рік тому

    Thanks A LOT for explaining why the test statistic follows the chi-squared distribution. It's so much better than just learning the formulae¬

  • @kalyanakrishna6524
    @kalyanakrishna6524 3 роки тому +1

    Awesome explanation! Especially about the hidden z-distribution. Thanks a ton!!

  • @deanemilne6275
    @deanemilne6275 2 місяці тому

    nicely explained. you answer all the questions that other videos don't address

  • @rishabhchoudhry5618
    @rishabhchoudhry5618 4 роки тому

    Finally understood where the normal distribution fits into the test statistic. Thanks a lot

  • @danvachidoh1846
    @danvachidoh1846 4 місяці тому

    Thanks much for this explanation. you real nailed it, specifically on obtaining observed and expected frequencies.
    Previously I used hours and hours without understanding. I am real happy to know this

  • @sahilkathuria9850
    @sahilkathuria9850 Рік тому

    Thank you for these great informative videos. Best explanation I found till now related to the Chi square distribution.

  • @louisdedeken2904
    @louisdedeken2904 5 років тому +4

    Absolute legend, love your work m8te, keep up the good work
    Cheers from Belgium

  • @rajat4640rajat
    @rajat4640rajat Рік тому

    Nice one. I was really searching for the explanation of the expression used in the test, which I got at the end of your video. Thank you so much.

  • @xinfeili6211
    @xinfeili6211 3 роки тому

    Your video is so detailed along with our curiosities! My mind got clear through this easy and fun video.

  • @elvinphilip3410
    @elvinphilip3410 2 роки тому

    You are a gifted teacher! Thanks!

  • @iftrejom
    @iftrejom 2 роки тому

    Thank you so much for such a comprehensive guide to chi-squared. Yo u made it so easy to understand!

  • @harshitmangwani7213
    @harshitmangwani7213 3 роки тому

    Tons of gratitude for this video. Easily one of my all-time favorites :)

  • @adityasatyam6300
    @adityasatyam6300 3 роки тому

    This is brilliant Sir . YOU ARE A GEM really .

  • @cococnk388
    @cococnk388 2 роки тому

    Concepts explanation make easier to understand the formula.
    Anyway I have never been able to memorise a formula..
    Thanks Zed.

  • @rajvimittal3753
    @rajvimittal3753 2 роки тому

    This was just amazing. I got it in one go! Thanks a lot for doing this!

  • @Ai-ChingChen
    @Ai-ChingChen 2 роки тому

    Such amazing explanation that just about to make me cry 😭

  • @manog8713
    @manog8713 10 місяців тому

    Excellent presentation and explnantion. Well done.

  • @akashdatta1820
    @akashdatta1820 7 місяців тому

    Thanks a lot. I wish I could depict how much you had helped me.

  • @haraldurkarlsson1147
    @haraldurkarlsson1147 8 місяців тому

    Very nice demonstration and explanation. However, excel is not the only we to calculate the values for chi-squared. R does an excellent job and is free to boot.

  • @JiaqiDing
    @JiaqiDing Місяць тому

    Perfect tutorial video! Thanks ❤

  • @zhizhongpu8937
    @zhizhongpu8937 3 роки тому +2

    Since the idea comes from a binomial distribution, why do we use a Chi-squared test, instead of developing some tests based on binomial distribution?

  • @PunmasterSTP
    @PunmasterSTP 7 місяців тому

    Goodness of fit? More like "Great lectures that are lit!" 🔥

  • @LanaUSA212
    @LanaUSA212 4 роки тому +1

    Thank you so much ❤️ I am in grad school and your videos are the life savers. My college professor can not explain everything as good as you do !!!!

  • @Engineer78206
    @Engineer78206 6 місяців тому +1

    great video

  • @jameszhai6160
    @jameszhai6160 2 роки тому +1

    Hey Justin, An awesome explanation about chi-square: I had read lots of articles and lectures about chi-square expecting to know why it can be used on goodness of fit tests, but no one gave explanation about why. You explained well using binomial distribution. Thank you, and will follow you on statistics topics.

  • @Competent00
    @Competent00 3 роки тому

    I never knew it would be this easy until i watched this video

  • @newtonocharimenyenya2458
    @newtonocharimenyenya2458 3 роки тому

    A Very GREAT Piece.

  • @appdroide100
    @appdroide100 4 роки тому

    Really cool video bro, it's absolutely well explained , keep it up!

  • @dilaraesmer
    @dilaraesmer 3 роки тому

    "Learning to predict rare events from sequences of events
    with categorical features is an important, real-world, problem that existing statistical and machine learning methods are not well suited to solve." Gary M. Weiss*
    and Haym Hirsh
    Thank you for your video :)

  • @sinaabedini7487
    @sinaabedini7487 4 роки тому

    Beautifully video, skillfully made. Thank you

  • @oishaneekar8024
    @oishaneekar8024 4 роки тому

    you deserve a lot more!!!!

  • @MDMAx
    @MDMAx 2 роки тому

    Thank you, teacher!

  • @josechvaicer7328
    @josechvaicer7328 3 роки тому +1

    Excellent video! My question, if you may, is why for the 2nd example the complementary deviations are not included in the Summation formula ? What I mean is for the 1st example the left hand + the right hand deviations were summed up, but for the 2nd example, you included only those who choose ROCKs + Only those who choose scissors and only those who choose paper, but not those who did not choose Rocks + those who did not choose scissors + those who did not choose paper. Why is that so?

  • @truongphu7407
    @truongphu7407 3 роки тому

    It is helpful video. Could you tell me how to find the p value range (a

  • @deepblueocean5055
    @deepblueocean5055 2 роки тому

    This is really informative, Thank you!

  • @nick45be
    @nick45be 3 місяці тому

    21:25 where Is the link?

  • @dadmehr6841
    @dadmehr6841 2 роки тому

    Great video! Thank you!

  • @avisharma5394
    @avisharma5394 3 місяці тому

    super good explanation thanks

  • @Dupamine
    @Dupamine 9 місяців тому

    But why do we specifically use chi square to do this fit test ? Why not any other distribution such as exponential ? Why specifically chi square ?

  • @carlostolosa6530
    @carlostolosa6530 4 роки тому

    8:55 why k=1 but we have 2 terms in the summation?

  • @matomerabothata1526
    @matomerabothata1526 4 роки тому

    you are excelling on this sudject

  • @ashraf736
    @ashraf736 3 роки тому

    Thanks for the knowledge

  • @manuelolazabalmoran2667
    @manuelolazabalmoran2667 3 роки тому

    Why Chi-Squared test is used for qualitative variables if the distribution is obtained from a Gaussian distribution (which is quantitative continuous data)?

  • @WandaMusa-o1h
    @WandaMusa-o1h Рік тому

    It's a good video you have really helped me

  • @jmpoul0181
    @jmpoul0181 4 роки тому +1

    Tres bien explique.

  • @programmer1010
    @programmer1010 2 місяці тому

    So when dof=1 chi-squared goodness of fit test is same as testing for population proportion?
    Because when dof=1, chi-squared at alpha significance level is z squared at alpha significance level

  • @adamkinsey3139
    @adamkinsey3139 3 роки тому

    Using the Binomial distribution, the answer you get to this question is that there's a 10.2% chance of getting this data from such a sampling.
    (assuming that the "true" percentage of left-handers in society is 12%.)
    Why does using the Binomial method give such a drastically different result than the Chi-squared test method?

  • @luiscalama8737
    @luiscalama8737 3 роки тому

    Nice video, it helps a lot

  • @sheetalborar6813
    @sheetalborar6813 2 роки тому

    i love your voice!

  • @blkscreen15
    @blkscreen15 10 місяців тому

    Since this is a two-tailed problem, should the alpha be 2.5%?

  • @Vivekagrawal5800
    @Vivekagrawal5800 2 роки тому

    Is that z similar as z distribution or z test formula?

  • @zhuwenhao4852
    @zhuwenhao4852 Рік тому

    nicely done.

  • @7justfun
    @7justfun 3 роки тому

    How would you set Ho to be , shouldn't we be setting Ha to be 12% because that's what we want to prove ?

  • @kristendang6778
    @kristendang6778 7 місяців тому

    How do you get the level of significance at 5%? Is this standard?

  • @paschaliskagias9675
    @paschaliskagias9675 2 роки тому

    Hi. Your videos are great. I have one question. Chi-squared is appropriate to be used to test the first digit law (Benford's law)?

  • @samueldarenskiy6893
    @samueldarenskiy6893 Рік тому

    Wouldn't you need to do a two tale test for this?

  • @jivillain
    @jivillain 4 роки тому +1

    20:23 n*pi > 5; n(1-pi) > 5. Why is that?

    • @powerclasses2931
      @powerclasses2931 4 роки тому +1

      The Central Limit Theorem applies even to binomial populations like this provided that the minimum of np and n(1-p) is at least 5, where "n" refers to the sample size, and "p" is the probability of "success" on any given trial. Probabilty of Success here is proportion of population left handed = 0.12

  • @tomharte9371
    @tomharte9371 4 роки тому

    Excellent video.
    How does the sample size flow through from the Normal distribution into the Chi distribution? If your example had 110 left handed students out of 750 in the art class then we would reject the null hypothesis?

  • @sanchitamayekar7247
    @sanchitamayekar7247 3 роки тому

    This is amazingg!! Thank you so muchh!!

  • @ShubhamSingh-lx6fd
    @ShubhamSingh-lx6fd 2 роки тому

    How did u get 200 in Excepted row for rock paper scissors?

  • @loicvandenbroeck2532
    @loicvandenbroeck2532 5 років тому +1

    Hi ! Thanks for the video, will you make another on the Fisher distribution ?

  • @jonathanwieler8863
    @jonathanwieler8863 3 роки тому

    Thank you!!! This helps a lot!!!

  • @vrushalibhise7375
    @vrushalibhise7375 Рік тому

    Could you pls pls pls help me with the solution of this qn...
    A frequency data is classified in 9 classes and Gamma distribution is fitted to it after estimating the Parameters. If a Chi square, goodness of fit test is to be used without combining the classes, the degrees of freedom associated with chi square test are:
    9
    8
    7
    6
    Could you help me answer this!

  • @loislovescheese4776
    @loislovescheese4776 Рік тому

    what if you had 3 replicates for each variable (say i have variables a, b, c, d and 3 replicate values for each (eg variable a has values 0.1, 0.2, 0.3) can I do the chi squared test on the 4 variables? Do I have to find the mean of each variable then do the chi squared test? or is there a better test for that? Thanks!

  • @janinajochim1843
    @janinajochim1843 4 роки тому +1

    Thanks for the video!!!!
    One quick question that's beyond me:
    Once we know that a binomial distribution with a large n is basically a normal distribution (this explanation may be clumsy) why WOULD we square is. In the video you are saying "What happens if we square it?" but what is the reason for squaring it ?? Thank you!!!

    • @adarsharao
      @adarsharao 4 роки тому

      I am not as good as him in explaining.. check out ua-cam.com/video/80ffqpZdKiA/v-deo.html 10:00

  • @pedrodasilva9189
    @pedrodasilva9189 2 роки тому

    Hello Zed, could you please do a video on different hypotheses testing for contingency table models? Thanks!

  • @ellenhart2841
    @ellenhart2841 3 роки тому

    Thx. Great Video!

  • @uzmaiqbal1952
    @uzmaiqbal1952 4 роки тому

    Thank you soo much for lecture this topic kindly upload lecture about basian or base theorem

  • @weeradasattayawuthipong3460

    thank you so much

  • @fatriantobong2097
    @fatriantobong2097 3 роки тому

    why dont we split the alpha by 2?

  • @tensor131
    @tensor131 3 роки тому +1

    Really great stuff this. I'm working my way through your chi-squared vids. Now have I missed something here? If you give that question to a student who has not yet met chi-squared, surely it is solvable by using a binomial hypothesis test (H0: X~B(75,0.12) etc.) I guess my question is this: 'in the real world' which method do statisticians use for 2 category data and does it always reach the same conclusion? It would seem odd to opt for a test that uses an approximation (large n normal approx to binomial and/or approximation to chi-squared) when one can easily (using a computer or advanced calculator) carry out an EXACT binomial hypothesis test. I can see that for more than 2 categories, you would need a "multinomial hypothesis test" (if such a thing exists) and that sounds way more complicated than chi-squared in that case!

  • @subtlethingsinlife
    @subtlethingsinlife 2 роки тому

    Still I am not able to intuitively understand as to why chi square distribution can be used for testing goodness of fit .. still can't wrap my head around

  • @IQBRILLIANT-sj4ys
    @IQBRILLIANT-sj4ys 5 місяців тому

    Is it not two tail ?

  • @grouse6
    @grouse6 4 роки тому

    Can you use chi squared on outcomes of an intervention. For example if the results are positive or negative.

  • @kanishkbakshi10
    @kanishkbakshi10 4 роки тому

    I performed a chi-square goodness of fit test on my Signal Detection Theory. In the residuals, I am seeing that the residuals of Hits and Misses are mirror images of each other (For example: 1 and -1) and similarly for the residuals of False Alarms and Correct Rejections, they are also mirror images of each other (For example: 5 and -5). I used counterbalancing while collecting data. Can you think of a reason as to why I am seeing these mirror images? Also, do you know of any references (papers, articles etc) that I can go through.

  • @ricardoafonso7563
    @ricardoafonso7563 3 роки тому

    .
    Thank you
    A nice lesson..
    at age nearly 60 ...soon
    .

  • @kristalkristal2506
    @kristalkristal2506 4 роки тому

    When you use the number pi do you mean p as in probability or the irrational number pi? I don't mean to nitpick, but this is confusing to read since pi is like a reserved word in mathematics. I've never seen the letter pi being used to refer to anything other than the constant pi, and seeing it here, I really have to think about what it is that you intend to express.
    I'm honestly confused about the formula at 21 minutes because (1-pi) is negative, and there is no real square root of a negative number.
    Help, please?

    • @zedstatistics
      @zedstatistics  4 роки тому

      Believe it or not, in statistics we often use the greek letter pi to mean the "population proportion". I mention that this is what it represents at 18:25. Hope that helps :)

  • @sciencescience9102
    @sciencescience9102 4 роки тому

    Super!!!!! Thank you!!

  • @vinodkumarreddyreddy353
    @vinodkumarreddyreddy353 3 роки тому

    Hi, I have seen both of your videos on Chi-square goodness of fit and Test for independence and got confused as I found both solutions and conclusions to be similar except for p-value could you say in brief what is the main difference between both the Tests.

    • @imanesarrou6708
      @imanesarrou6708 7 місяців тому

      Hello there did you find the answer to your question cz I have the same concern and I still can't figure out how to differentiate between the two methods. I would appreciate it if you could provide me with an answer.

  • @youcefyahiaoui1465
    @youcefyahiaoui1465 3 роки тому

    Zedstats, awesome course. However, right about 16:28 in the video you keep saying that the p value is the shaded area. It is actually not. p

  • @joeo1598
    @joeo1598 3 роки тому

    Is there such thing as confidence intervals for our proportions, with Chi squared? Or is that only doable with parametric stats?

  • @benjaminmoshref1601
    @benjaminmoshref1601 4 роки тому

    Thank you so much, I finally understood how to do a goodness of fit test after struggling with it for more than a week. But I've got one question: if I'm applying it to a linear fit of a given set of physical measurements, so I have a one by one relationship between the expected values and the observed ones, can I do a chi square test? Because one of the applicability hypotheses for doing this kind of test is to exceed at least the frequency of five but for a linear fit what does it mean? Do i have to kind of group my data or I am not able to do this chi square test in the case of a linear fit? Thank you in advance, I am a second year physics student

  • @claudio_rosa.consultoria5074
    @claudio_rosa.consultoria5074 2 роки тому

    Fantastic :)

  • @henrmota
    @henrmota 4 роки тому +1

    how about to do an entire course with quality and get the payment you deserve, as we get the knowledge we deserve.

  • @lukavidakovic2385
    @lukavidakovic2385 4 роки тому

    thanks for the help

  • @marioalberto10
    @marioalberto10 2 роки тому

    WHERE DO YOU GET THE EXPECTED VALUE IF NOT FROM AN OBSERVATIONAL STUDY, I MEAN WTF

  • @LetsgoWw3
    @LetsgoWw3 4 роки тому

    Thanks a lot