Session 43 - Central Limit Theorem | DSMP 2023

preview_player
Показать описание


--------------------------------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------------------------

-------------------------------------------------------------------------------------------------------------------------------------------------------

-----------------------
| **Chapters** |
-----------------------
00:00:00 - Session start
00:03:21 - Agenda of this session
00:05:35 - Bernouli Distribution
00:15:31 - Binomial Distribution
00:22:42 - Solve some problems using binomial distribution
00:27:26 - PDF formula for binomial distribution
00:31:06 - Graph of PDF for the binomial distribution
00:33:36 - Binomial Distribution in Code
00:39:38 - Criteria of the Binomial Distribution
00:40:32 - Applications of Binomial Distribution in Data Science
00:45:35 - Sampling Distribution
00:54:32 - Why Sampling Distribution is important?
00:57:05 - Intuition of Central Limit Theorem (CLT)
01:06:57 - Central Limit Theorem in code
01:14:08 - Case Study 1 - Fare of Titanic Dataset
01:32:00 - Doubt clearance
01:37:51 - Assumptions of making samples
01:38:33 - Case Study 2 - What is the average income of Indians?
01:42:27 - Session end & Doubt clearance

#datanalytics #Stats #statistics #SQL #descriptivestatistics #campusx #dsmp2023
Рекомендации по теме
Комментарии
Автор

you lectures will remembered as a masterpiece even after 10 years from now

shashankshekharsingh
Автор

I have a doubt that why at 1:30:32 we divided sample standard deviation with sqrt of sample size when the formula is sample s.d = (population s.d) / (sqrt of sample size).

NikhilKumar-jyzy
Автор

each and every point is clear sir. Thank you sirji.

mayanksinghrana
Автор

I have a small correction to make at 27:45, The formula is actually of PMF( Probability Mass Function) since Binomial Distribution is a discrete distribution, hence it is PMF and not PDF (Probability Density Function) which is used for continuous random variable.

kameshkotwani
Автор

Sir, Your a Gem, anyone can lecture but only few can teach and make class interesting, my gratitude towards you will be forever, thank You for this session .
Please keep Teaching, Mentoring . ❤🙏🙏🙏

KamaleshgowdaOfficial
Автор

Sir please 😊 bring videos on Time series analysis using ARIMA and SARIMAX Models

SameerAli-nmxn
Автор

sir love you from pakistan sir you're truy gems for machine learning students sir tysm for provide us this content tysm sir

AliAhmadae
Автор

this lecture was fucking awesome. I understood confidence interval which was very difficult for me until now. Is there any way I can donate you some money? not by being a member on Paypal or something? please let me know.

prathameshchaudhari
Автор

Hi Sir, We have to do sample_means.std() * np.sqrt(50). That is why the error is occurring. Since the sampling std deviation is 1/sqrt(n) then actual standard deviation will be * sqrt(n).

Iknownothink
Автор

1:50:07 I think chat me log ye puchh rahe the ki it should be
Pop.mean / √n, but you have written sample.mean /√n.

If sample mean is pop.mean/√n already then why would you divede SAMPLE.mean /√n? Which makes pop.mean/√n/√n twice divide by √n...

jpatel
Автор

Great masterpiece, rare to see a very smart and hardworking person in teaching, the best thing is you put a lot of effort in developing intuition, most teachers don't even know what intuition is

parthmishra
Автор

Sir, you have used the entire population to calculate the sample mean which introduces bias in sampling. It would be better to do it on 40% sample to demonstrate the power of this theorem

shreyammisra
Автор

1:07:40 Why do sample means vary in case of uniform distribution? In uniform distribution, each point will have the same probability, so no matter whatever sample you create each sample will have the same sample mean which will be equal to the population mean, which will just create a single point, and that point will be the population mean in the sampling distribution of sample means isn't it? or I'm missing something here?

Edit : after playing around with it I figured out that here we are talking about probabilities being same, but values are different, and our mean in this case is expected values, since values are different, their probability density will be same for each sample but sample mean will vary. this is why we will get the different sample means in sampling distribution.

dhruvilpatel
Автор

itne easily koi nahi samjah sakta... u r a miracle🙏🙏🙏🙏

shivika
Автор

Sir this is just incredible! What great content really sir !

rushikesh
Автор

this is pure gold and now i m becoming a member

c.nbhaskar
Автор

Hello sir, I can't find any videos on uniform and poisson distribution of discrete probability distribution in your channel . Have you uploaded them or not ?? It is not in the maths for machine learning playlist

nitinbajaj
Автор

Thank you for this amazing session! You are really a best teacher!

researchburst
Автор

Sir really CLT outstanding explanation tha pura hi visualise hua🤜🏻🤛🏻

-vishalyadav
Автор

20:50 The probability of getting like on your video is 1 from my end.

romanojha