Tutorial 17-Hypothesis Testing And Statistical Analysis Using Z test

A Z-test is any statistical test for which the distribution of the test statistic under the null hypothesis can be approximated by a normal distribution. Z-tests test the mean of a distribution.
Support my channel and join the membership so that I can uplaod more videos
/ @krishnaikhindi
--------------------------------------------------------------------------------------------------
Check out our Full Stack Data Science Masters Course With Job Assistance
bit.ly/3UGeZQz
------------------------------------------------------------------------
All Playlist links are given below
NLP Playlist: • Natural Language Proce...
ML playlist in hindi: bit.ly/3NaEjJX
Stats Playlist In Hindi:bit.ly/3tw6k7d
Python Playlist In Hindi:bit.ly/3azScTI
---------------------------------------------------------------------------------------------------------------
Please donate if you want to support the channel through GPay UPID,
Gpay: krishnaik06@okicici
Telegram link: t.me/joinchat/N77M7xRvYUd403D...
-------------------------------------------------------------------------------------------------------------
Please join as a member in my channel to get additional benefits like materials in Data Science, live streaming for Members and many more
/ @krishnaik06
-----------------------------------------------------------------------------------------------------------
Please do subscribe my other channel too
/ @krishnaikhindi
---------------------------------------------------------------------------------------------------------
Connect with me here:
Twitter: / krishnaik06
Facebook: / krishnaik06
instagram: / krishnaik06

Пікірлер: 106

  • @vyasishan3800
    @vyasishan38004 ай бұрын

    Thank you Krish for putting efforts to explain this concepts.

  • @arpitasharma28
    @arpitasharma287 ай бұрын

    Really appreciate your efforts to upload this video... great explanation sir:)

  • @amartamboli6718
    @amartamboli671811 ай бұрын

    Null Hypothesis = 5 Alternative Hypothesis = not equal to 5 Significance level = 2%,1-1% either side... Area under curve using positive z-score table = 1-0.01=0.99=2.33(As per Z-score positive table) Decision Boundary = +- 2.33 Z-score = -2.53 Since Z-test value is less than -2.33 i/e. it is in the negative rejection region , So we reject the null Hypothesis ..

  • @salmanshaikh5851

    @salmanshaikh5851

    11 ай бұрын

    Bhai mere ko to abhi bhi palle nai pada Kaha se tune sikha batai gake?

  • @hades840

    @hades840

    6 ай бұрын

    it is one tail test look there is underline on less

  • @harshshukla980

    @harshshukla980

    Ай бұрын

    decision boundary 2.05

  • @sarveshpandey9047

    @sarveshpandey9047

    Ай бұрын

    2.33 kha se aya z table me to diya he nhi h

  • @rohitrohit6883
    @rohitrohit68839 ай бұрын

    Sir, I am very thankful to you for this video.🙂😇

  • @MuhammadIbrahim-ju1tp
    @MuhammadIbrahim-ju1tp2 ай бұрын

    great krish respect so much value

  • @vishaltiwaribnt
    @vishaltiwaribnt6 ай бұрын

    thanks for Uploading in Hindi.....its much more understaniding.....

  • @MH_POLITICSS
    @MH_POLITICSS8 ай бұрын

    Decision boundry= +-2.33 Z score = - 2.52 Conclusion - if z score is less than -2.33 or greater than -2.33 then we reject null hypothesis. So -2.52 >-2.33 we Reject null hypothesis..

  • @anuraaganath
    @anuraaganath10 ай бұрын

    pmean = 5, std = 0.5, smean=4.8, n=40; null hypothesis = pmean = 5 alternate hypothesis = pmean C.I = 1-0.02 = 0.98 z = -2.52 where the decision boundry is -2.06 as per the z table. So, we failed to accept the null hypothesis and the warranty should be revised

  • @alex-vq1yy

    @alex-vq1yy

    3 ай бұрын

    can you explain how decision boundary = -2.06 how you calculated CI please explain

  • @ashan3453
    @ashan34536 ай бұрын

    the explanation is too good sir

  • @mahalaxmiindapure8961
    @mahalaxmiindapure89615 ай бұрын

    I got decision boundary between -2.33 to +2.33, z-score= -2.529, We will reject the null hypothesis means workers believe is true.

  • @sumitkoshti1171

    @sumitkoshti1171

    3 ай бұрын

    Yes

  • @madhuchandra123
    @madhuchandra12311 ай бұрын

    Please complete the statistics playlist.

  • @ashishvinod2193
    @ashishvinod21937 ай бұрын

    since our entire area under the curve value is = 1 -0.98 = 0.02 since it's left tail test.. out z test value =-2.52 2.52

  • @rajsinghtomar8090

    @rajsinghtomar8090

    4 ай бұрын

    no confidence interval how i solve plz reply

  • @ashishvinod2193

    @ashishvinod2193

    4 ай бұрын

    @@rajsinghtomar8090 if confidence interval was not given then by default assume 98% confidence intervals

  • @sumitkoshti1171

    @sumitkoshti1171

    3 ай бұрын

    @@rajsinghtomar8090 bhai significance value is given 2% which is equal to 0.02, and by using significance value you can find CI from the formula significance value = 1-CI, 0.02 = 1-CI, CI = 1-0.02 = 0.98 which is 98%

  • @sketchytv1321
    @sketchytv132111 ай бұрын

    بہت آسانی سے مشکل کنسپٹ سمجھایا ۔شکریہ

  • @EjazUddin786

    @EjazUddin786

    3 ай бұрын

    Hanji bilkul

  • @shaitan779gaming9
    @shaitan779gaming915 күн бұрын

    Hi Krish, is it standard that we should always go with Z-test when sample size is above 30?

  • @hardikanandani7377
    @hardikanandani737711 ай бұрын

    Ho = average warranty is 5 years. H1= average warranty is not equal to 5 years. Alpha = 0.02(Here it's less than 5 years ,so it's one tailed.) Area under curve = 1-0.02 = 0.098 Decision boundry = +-2.06 Z-score = -2.52 So Ho is rejected. So, the worker was right I think. Bulb will malfunction in less than 5 years.

  • @souravsinha18

    @souravsinha18

    11 ай бұрын

    yes it is correct but decision boundary should be -2.06 not + - 2.06

  • @hardikanandani7377

    @hardikanandani7377

    11 ай бұрын

    Correct.

  • @sumiturkude1395

    @sumiturkude1395

    11 ай бұрын

    @@hardikanandani7377 how did you find decision boundary.?

  • @shashi_kamal_chakraborty

    @shashi_kamal_chakraborty

    8 ай бұрын

    @@sumiturkude1395 It's 1 - alpha = 1 - 0.02 = 0.98. Now, z-table look up for 0.98. Which is found to be -2.06

  • @KartikSharma12364

    @KartikSharma12364

    4 ай бұрын

    A chocolate company claims that their new recipe for dark chocolate bars increases the average cocoa content from the previous recipe. They conducted a study where they sampled 50 dark chocolate bars and found that the sample mean cocoa content is 85%. The population standard deviation of cocoa content is known to be 8%. Test the company's claim at a significance level of α = 0.05. Solve this problem

  • @zarmeenakhan899
    @zarmeenakhan8995 ай бұрын

    can you make a video on cohort and case control as well?

  • @salmanshaikh5851
    @salmanshaikh5851 Жыл бұрын

    Continue Sir Continue me sahi time per aaya hu bcz mera abhi 15 ma tutorial chalra hai😂🤣

  • @Khojbysridhar
    @Khojbysridhar14 күн бұрын

    You are amazing

  • @KartikSharma12364
    @KartikSharma123644 ай бұрын

    A chocolate company claims that their new recipe for dark chocolate bars increases the average cocoa content from the previous recipe. They conducted a study where they sampled 50 dark chocolate bars and found that the sample mean cocoa content is 85%. The population standard deviation of cocoa content is known to be 8%. Test the company's claim at a significance level of α = 0.05. Anyone solve

  • @user-vb5hh7rw3s
    @user-vb5hh7rw3s9 ай бұрын

    for decision boundry i use this : for 2% significane 1-1 on both sides 1/100=0.01 , 1-0.01=0.99 and in z table value of 0.99 is 2.33 means +2.33 on right and -2.33 on left pls tell me am i correct ?

  • @rajatjuneja8367

    @rajatjuneja8367

    4 ай бұрын

    yes sir

  • @rishabhchauhan8612
    @rishabhchauhan861211 ай бұрын

    This topic is For Data Analytics or Data science ??

  • @rajarshiroy8794
    @rajarshiroy87948 ай бұрын

    I have a doubt, From other definitions I think that If we have the population standard deviation we can directly move for z test since we are concerned about the sample size only when we are talking about the sample standard deviation as for n>30 sample and population SD are similar. So, I think the flowchart would be a little different. Can anybody clarify this doubt please??

  • @harshitgupta6452

    @harshitgupta6452

    28 күн бұрын

    Yes you are correct

  • @baishalibiswas7525
    @baishalibiswas75256 ай бұрын

    Please complete full playlist

  • @Vishal_19
    @Vishal_192 ай бұрын

    U = 5 sd = 0.50 n = 40 x bar = 4.8 so area = 1 - 0.02 = 0.98 z_value will be 2.06 (dission boundry) 4.8 - 5 ----------- 0.05 * root 40 = - 2.52 we reject the null hypotheses the worker was right

  • @sauravjha1714
    @sauravjha171411 ай бұрын

    Null Hypothesis = Mean = 5 Alternative Hypothesis = Mean not equal to 5 Significance level = 2%,1-1% either side... Area under curve using positive z-score table = 1-0.01=0.99=2.33(As per Z-score positive table) Decision Boundary = +- 2.33 Z-score = -2.52 Since Z-score falls in the negative rejection area...we reject the null Hypothesis!!!

  • @rokhaiyasultana9728

    @rokhaiyasultana9728

    11 ай бұрын

    i got the sme ans!!

  • @souravsinha18

    @souravsinha18

    11 ай бұрын

    No it is not a 2 tailed test so alternative hypotheis should be mean less than 5 years

  • @salmanshaikh5851

    @salmanshaikh5851

    11 ай бұрын

    Bhai mere ko to abhi bhi palle nai pada kaha se tune sikha pata gake?

  • @sauravjha1714

    @sauravjha1714

    11 ай бұрын

    @@souravsinha18 check your answer... My answer is perfectly right!!

  • @souravsinha18

    @souravsinha18

    11 ай бұрын

    @@sauravjha1714 no bro I did this I am sure this is right why are u taking 1-1% on both sides when the worker says that the bulb lifetime will decrease you should take it on decreasing side and then calculate it

  • @manjeshtiwari7434
    @manjeshtiwari74345 ай бұрын

    Thank you for such an amazing lecture

  • @KartikSharma12364

    @KartikSharma12364

    4 ай бұрын

    My decision boundary is not getting correct ,can you help me

  • @manjeshtiwari7434

    @manjeshtiwari7434

    4 ай бұрын

    @@KartikSharma12364 what's issue?

  • @manjeshtiwari7434

    @manjeshtiwari7434

    4 ай бұрын

    @@KartikSharma12364 if accept area is 98 how can be reject 49 and 49, out of 100 98 is accept area so we have 1% reject area on both side u need to rewatch

  • @KartikSharma12364

    @KartikSharma12364

    4 ай бұрын

    @@manjeshtiwari7434 what's the correct answer of decision boundary and z test

  • @KartikSharma12364

    @KartikSharma12364

    4 ай бұрын

    ​@@manjeshtiwari7434decision boundary=2.33

  • @AmitSharma-oh5uw
    @AmitSharma-oh5uw7 ай бұрын

    make more video correlation regression and probability in stats

  • @user-uh6th9vc5x
    @user-uh6th9vc5x7 ай бұрын

    V nice Sir

  • @HarshitSharma-df7kx
    @HarshitSharma-df7kx18 күн бұрын

    sir what do we do when we do not have a 2 tail test condition ??

  • @shailendrayadav6937
    @shailendrayadav69375 ай бұрын

    Here bulb manufacturers question is a one tail test question because , worker believe that bulb will malfunction in less than 5yrs . So Null hypothesis= H0 = 5yrs alternate hypothesis = H1

  • @xero4479
    @xero44792 ай бұрын

    Sir kal paper hai aur ye pura samajh hi nahi aa raha tha, thank you🙏

  • @dhananjaykharat4812
    @dhananjaykharat4812Ай бұрын

    (Z Test = -2.53)H0 has been rejected as (Z test VAlue is not in C.I range/Decision Boundry) so worker believe is correct.

  • @rajasaad3450
    @rajasaad345011 ай бұрын

    wait for course completion !

  • @swapnil2881
    @swapnil288111 ай бұрын

    Sir please stat ki playlist continue rakho

  • @pr_ateek
    @pr_ateek9 ай бұрын

    Please complete the statistics playlist

  • @Cricstory0108
    @Cricstory01086 ай бұрын

    Sir konse video ha hypothesis check karne wali

  • @alphapoint389
    @alphapoint3894 ай бұрын

    Sir how formulas are derived?

  • @kdasurrenderedsoul7613
    @kdasurrenderedsoul76135 ай бұрын

    Hare Krishna 😊 Thank you so much for this vidoe 😊😊😊

  • @vaishhh_likhii
    @vaishhh_likhii7 ай бұрын

    Thankyou so muchh,such a nice explaination!! best video for Hypothesis testingg......🤌

  • @zero_analyst
    @zero_analyst9 ай бұрын

    Hello sir thank you for explaining You have mentioned in conclusion if z test is less than 1.96 or greater than 1.96 then we reject Null Hypothesis but as results came 2.31 which is greater than 1.96 so we should accept the Null Hypothesis. I am now confused can you please clarify thanks in advance

  • @itshimanshucreation8076

    @itshimanshucreation8076

    8 ай бұрын

    Bro unhone bola hai ki we accept to reject the H0 if z test is lessthen or greater then 1.96.

  • @vedchaudhari6587

    @vedchaudhari6587

    5 ай бұрын

    if it lies in range of 1.96 and -1.96 then accept it else for anything either greater or smaller , reject it.

  • @wewesusui9528
    @wewesusui952811 ай бұрын

    Sir anova test pr video banao baki ke saare test statistic 7 days live me hue he cover pr anova test nahi hua he

  • @heeralalchoudhary1587
    @heeralalchoudhary15879 ай бұрын

    Fabulous sir ..can I contact personally

  • @godgdfhackergod3836
    @godgdfhackergod383611 ай бұрын

    please continue video on ml

  • @dilipgyawali1776
    @dilipgyawali17764 ай бұрын

    if population standard deviation is given and sample size is less than 30, then Z test???please clarify.

  • @robinsunuwar6852

    @robinsunuwar6852

    3 ай бұрын

    you should go with t test if the sample size will be less than 30 we have to perform t test its basic rule.

  • @hardsoul00
    @hardsoul008 ай бұрын

    Sir CHI SQUARE AND ANNOVA KA VIDEO Bna dein mera paper ha iska next month 10th november ko

  • @dishantsonawane8510
    @dishantsonawane85102 ай бұрын

    Why do we consider sample size 30 for differentiating between Z test and T test. Why not some other number ? Why not 'x' percentage of the population ?

  • @raufurrahim878

    @raufurrahim878

    Күн бұрын

    30 data points should provide enough information to make a statistically sound conclusion about a population. This is known as the Law of Large Numbers, which states that the results become more accurate as the sample size increases. The 30 is a rule of thumb, for the overall case, this number was set by good statisticians.

  • @user-wn6tv1wm7u
    @user-wn6tv1wm7u4 ай бұрын

    Is This playlist enough to learn statistics for data analytics ?

  • @Prashantchaudhary-kf1fb

    @Prashantchaudhary-kf1fb

    4 ай бұрын

    yes bro

  • @vishujaiswal6160
    @vishujaiswal616010 ай бұрын

    next video pl. sir

  • @akashshakya8740
    @akashshakya87406 ай бұрын

    pls provide lecture notes

  • @urvashirai8441
    @urvashirai84414 ай бұрын

    If std not given and sample size more thn 30,thn which test is to be used?

  • @CrocksJam

    @CrocksJam

    4 ай бұрын

    if population std mean not given then we use t-test directly

  • @Liku108
    @Liku1087 ай бұрын

    Sir what about f test

  • @fehmeedaayoub6559
    @fehmeedaayoub655911 күн бұрын

    Paired and unpaired ka concept????

  • @muhammadansar3373
    @muhammadansar33739 ай бұрын

    According to my calculation HO is accepted

  • @theentertainmentworld3310
    @theentertainmentworld331011 ай бұрын

    Do any one get the z test value?? I'm getting my z test value which is +0.037 and decision boundary values as +3.1 and -3.1 , So i have to accept the null hypothesis Do anyone get any answer??? Sir am i correct??

  • @sumitkoshti1171

    @sumitkoshti1171

    3 ай бұрын

    Nope, Decision boundary is +-2.33 and Z test value is -2.53, As -2.53

  • @prashindu
    @prashindu5 ай бұрын

    For a two-tailed test with a 98% confidence interval, you would use 0.01 (or 1%) for the purpose of arriving at the z-score. This is because the remaining 2% is split equally between the two tails of the distribution, leaving 1% in each tail. So, you’d look up the z-score that corresponds to an area of 0.99 in a standard normal distribution table. However, for a one-tailed test with a 98% confidence interval, you would use 0.02 (or 2%) to find the z-score. This is because the entire 2% is located in one tail of the distribution. So, you’d look up the z-score that corresponds to an area of 0.98 in a standard normal distribution table. When in doubt, it’s generally safer to use a two-tailed test. This is because a two-tailed test is more conservative and does not assume a direction of the effect.

  • @importantforlifeyasharora9042
    @importantforlifeyasharora904220 күн бұрын

    Actuall fanda ya to kisi ko pata nahi hai ya wo samjhane me inko takleef hai. Maths bhi theory ki tarah padha rahe hai😂

  • @danielrichard7575
    @danielrichard757511 ай бұрын

    "promo sm" 🤗

  • @robotsj2775
    @robotsj277510 ай бұрын

    Worst tutorial for z test This should not take more than 7-8 mins. Pls don't go into so much detail. Instead you would have solved 4-5 problems in 25 min.

Келесі