Hey great job on the video, I just have one recommendation. It probably would have been a good idea by showing what the area under the exponential distribution graph means and why your doing the 1-e^ thing. Like I learn better visually, so if I saw that P(X>value) means the area under the curve shaded from that value to infinity , it would better help me understand the probability.
Dave, I'm a bit confused here because i have been taught in the university that the expression for the Probability mass function is given by (Le^-Lx) where L stands for lambda or theta. But what you used in your video for the calculations is quite different
The formula in your comment is used as the pdf formula (probability density function) which calculates the probability P(X=x) , whereas P(Xx) is modeled using the cdf formula (cumulative one) - the one Dave used. Also a thing worth mentioning is that since X is a continuous random variable the probability that X will take an exact small x (some exact time) is virtually zero. Can you guarantee with absolute precision that something will happen at the exact time down to the very last 1^-inf second? No you can't. So every pdf formula for a continuous function you are taught actually calculates the probability that X will take a value in some interval, and that interval is just set in a way that it is theoretically converging to a single point to approximate the valid probability. So in a way , when X is a continuous random variable the cdf formula is used whenever you have some calculation about a concrete interval that you would want to calculate the probability for X lying in that inverval, while the pdf(Le^-Lx) should be used whenever you want to calculate the probability that X will take ~ some concrete value (the small interval whose width theoretically converges to zero that we talked about). Hope this clears up stuff.
Nice video but you should have mentioned that the graph is of λe^(-λt) and you just need to integrate to get the probability wherever you want and not remember all the jargon.
This is a great video, but I think the explanation of the difference between Poisson and Exponential was a little unclear. The way you explained it I thought the Poison formula would tell us the number of events that happen in an interval of time rather than that being the way the questions are posed. Overall this was a very helpful video. I've been starveling with identifying distributions and can't find a good video that breaks it down in this way. Thank you. If you made another video simply explaining how to identify all the major distributions I know it would be tremendously helpful to other students struggling through statistics.
Good video. I do wonder how textbooks come up with their problems. If German supermarktet clerk could only process 6 customers per hour, he or she would be a out of a job quickly.
I have another problem and I'd like to compare my approach based on your teaching of exponential dist you gave... I have a breakdown mean of 42 days and want to know the P of 34 days. For this problem, I assume I need to calculate the rate first which would be 8.69 breakdowns per year (Lamba = 8.69) .. then work out the P of
You are complicating the problem. You are given the expected value EX=42 days. To find the rate, you inverse EX, so lambda=1/EX=1/42 event will be done in average in a single day. Next, you are tasked to find P(X34) where X is the time taken for an event to happen modeled with Exp distribution X~E(1/42). Essentially the question can be rewritten as , find the probability that X will take any value EXCEPT 34. So all you have to do is subtract from 1 the probability P(X=34). Thus answer is 1- (1/42)*e^(-34/42) = 0.9894
Nice observation! The lambda you are talking about is the Poisson distribution lambda, which is how long on average for one event to occur. The lambda I use is another one: How many events occur in one unit of time. As you can see, these 2 types of lambdas are simply the inverse of one another. So your formula is correct as long as you use the Poisson lambda. Mine is also correct as I am using the inverse of your lambda. Hope this helps! David
thanks alot for this. But am really struggling with probability...i see it everywhere in staistics, it pisses me off. I dunno if you can suggest to me any book that is so self explanatory to explain these stats concepts or any link (preferably). Thanks Man
+Mario Lopez Hi Mario I guess you could call it intuition, or logic. What lies between 5 and 8? Well, everything below 8 minus everything below 5. Once you know this trick, when you see it again, it will be easy :) Good luck with your studies David
I think in a scientific calculator 'exp' means '10 rises to power of '...I think we should press simply the 'e' button instead of 'exp' button in the calculator..You can check this in online scientific calculator..
Hi Abdullah, thanks for your comment. I wasn't converting 6 hours to minutes. I was converting the number of customers we see per hour, to minutes. Which means, if we see 6 customers per hour, how many do we see in one minute? Do we see "6 x 60" customers per minute? Or do we see "6 / 60" customers per minute? Hint: It is the second one. We should be seeing LESS customers per minute.
there is an error in part B beacuse it says MORE than 10 minutes. this means finding the propability that the service will take 10 minutes similar to done in part A (said fewer than) and then taking the answer away from 1. had the question said less then it would be correct.
Look at the graph for Exp distribution variable, you will see that the highest probabilities are centered around 0 to the EX time taken. You can find the theoretical explanation on google.
If I understand you correctly , you are saying that he should be subtracting the interval (5, infinity) from the interval (0, 8)... isn't the result of doing that subtraction the interval (0, 5) intersect (8, infinity) which would indicate the number of customers attended to within the first 5 minutes and from 8 minutes till infinity - thus a deviation from the question?
Thank you so much. This video helped me tremendously in completing my homework!
Thanks, very to the point. Helped understand what tons of other videos couldn't
Hey great job on the video, I just have one recommendation. It probably would have been a good idea by showing what the area under the exponential distribution graph means and why your doing the 1-e^ thing. Like I learn better visually, so if I saw that P(X>value) means the area under the curve shaded from that value to infinity , it would better help me understand the probability.
Excellent. Concept plus practice examples. I loved it.
Thank you Dave for your video.
What book did you use?
Short and sweet. Thank you!
Dave you are the MAN - Legend, so easy explained... Thanks
Useful Video, thank you. I was just wondering, what is the textbook you are using? Thanks
It was definitely helpful, I got my doubt cleared!!! Thanks a bunch
hi - great video! thank you! can you tell us what is that book you used to show the problem please?
Great work, really legible and understandable
made it look so easy :) please help understanding the gamma and other distributions...
Brilliant video! Cheers for sharing it.
well explained!!..bravo
..... Save the planet!!!
You use a lot of papers
:(
Dave, I'm a bit confused here because i have been taught in the university that the expression for the Probability mass function is given by (Le^-Lx) where L stands for lambda or theta. But what you used in your video for the calculations is quite different
The formula in your comment is used as the pdf formula (probability density function) which calculates the probability P(X=x) , whereas P(Xx) is modeled using the cdf formula (cumulative one) - the one Dave used. Also a thing worth mentioning is that since X is a continuous random variable the probability that X will take an exact small x (some exact time) is virtually zero. Can you guarantee with absolute precision that something will happen at the exact time down to the very last 1^-inf second? No you can't. So every pdf formula for a continuous function you are taught actually calculates the probability that X will take a value in some interval, and that interval is just set in a way that it is theoretically converging to a single point to approximate the valid probability. So in a way , when X is a continuous random variable the cdf formula is used whenever you have some calculation about a concrete interval that you would want to calculate the probability for X lying in that inverval, while the pdf(Le^-Lx) should be used whenever you want to calculate the probability that X will take ~ some concrete value (the small interval whose width theoretically converges to zero that we talked about). Hope this clears up stuff.
You just explained something in 10 min that my professor could not in 1 hr 40 min.
sir when we could use the formula P(X>x) = e to the power of negative lambda multiply x and P(X
Hi my viewers! Are you in need of an online tutor? If so, check out the video description for details 😊
Thanks a lot, very helpful
Nice video but you should have mentioned that the graph is of λe^(-λt) and you just need to integrate to get the probability wherever you want and not remember all the jargon.
Excellent! Very well explained!
Thank you! Great tutorial :)
Dave saves the day! Thankyou!
Thank you for the help! My professor doesn't know how to teach and you definitely made this clear!
Amazing Teacher. Thank you for all you do.
Very useful video. Thanks tons
well it's gonna help me in exams it was useful,simple easy n fluent Thnx man...✌
Thank you, explained very well!
Thank u sir becoz of u i can do quiz, god bless you
This is a great video, but I think the explanation of the difference between Poisson and Exponential was a little unclear. The way you explained it I thought the Poison formula would tell us the number of events that happen in an interval of time rather than that being the way the questions are posed.
Overall this was a very helpful video. I've been starveling with identifying distributions and can't find a good video that breaks it down in this way. Thank you. If you made another video simply explaining how to identify all the major distributions I know it would be tremendously helpful to other students struggling through statistics.
Hey can u do some videos for normal distribution and other as well?
Thank u this was really easy to understand.
Thank you so much
Good video. I do wonder how textbooks come up with their problems. If German supermarktet clerk could only process 6 customers per hour, he or she would be a out of a job quickly.
Share with me that textbook you drew the example from
I have another problem and I'd like to compare my approach based on your teaching of exponential dist you gave...
I have a breakdown mean of 42 days and want to know the P of 34 days. For this problem, I assume I need to calculate the rate first which would be 8.69 breakdowns per year (Lamba = 8.69) .. then work out the P of
You are complicating the problem. You are given the expected value EX=42 days. To find the rate, you inverse EX, so lambda=1/EX=1/42 event will be done in average in a single day. Next, you are tasked to find P(X34) where X is the time taken for an event to happen modeled with Exp distribution X~E(1/42). Essentially the question can be rewritten as , find the probability that X will take any value EXCEPT 34. So all you have to do is subtract from 1 the probability P(X=34). Thus answer is 1- (1/42)*e^(-34/42) = 0.9894
Great Video! Thank You!
very clear, thanks a lot
Finally get it! Thank you!
Great tutorial, thanks!
What textbook were you using?
What about P(x < and equal 5)
it was really helpful. Thanks so much. Still struggling with probs and stats though.
Thank you. Well explained.
Very Informative!
Helpful 🙌
very easy to understand, thank u
thanks a lot!!! very useful
Hello Quant, can you tell the name of the book that you get this question? Thank you!
+Douglas Alencar UFMT Sure, its Statistics for Management and Economics, by Keller. I'm using the 9th edition.
+Quant Concepts Thanks man!
Hey Dave, what text book are you using in the video?
Thank you so much..
great explanation!
thanks a lot sir........its very helpful
what book did you refer to in the video? Thanks in advance
Fantastic!
Hi tks for the video. I have a question. (3:09) you wrote 1-e^(..). isnt it (x=0)+...(x=x)?
p(X>x)= 1-e^(..)? thank you
man you are awesome....THank you
Nice tutorial dude! thank you!
Life saver
Which book you are taking example from?
isn't the exponential distribution function supposed to be P(X > x) = e^(-x/lambda) and P(X< x) = 1-e^(-x/lambda), or is that something else?
Nice observation! The lambda you are talking about is the Poisson distribution lambda, which is how long on average for one event to occur. The lambda I use is another one: How many events occur in one unit of time. As you can see, these 2 types of lambdas are simply the inverse of one another.
So your formula is correct as long as you use the Poisson lambda. Mine is also correct as I am using the inverse of your lambda.
Hope this helps!
David
brilliant tutorials, but what about whiteboard?
Thank you so much, that was indeed helpful
sir, which text book u used in this video ?
it's really nice.
+ammireddy chirla Hi! I used "Business Statistics" by Keller. It's a classic :)
what is the book that u r using sir please????
I used "Business Statistics" by Keller.
what is the book name you read from it?
Thanks Bruv!
thanks alot for this. But am really struggling with probability...i see it everywhere in staistics, it pisses me off. I dunno if you can suggest to me any book that is so self explanatory to explain these stats concepts or any link (preferably). Thanks Man
Have a look at the Flaw of Averages by Sam Savage
How do you obtain that P(5
+Mario Lopez Hi Mario
I guess you could call it intuition, or logic. What lies between 5 and 8? Well, everything below 8 minus everything below 5.
Once you know this trick, when you see it again, it will be easy :)
Good luck with your studies
David
Quant Concepts Thanks!
thanks a million .
I think in a scientific calculator 'exp' means '10 rises to power of '...I think we should press simply the 'e' button instead of 'exp' button in the calculator..You can check this in online scientific calculator..
Hey, Please do answer my question. what tools are you using to record this video?
I'm literally using a handheld canon camera that is mounted on a flexible stand.
10:02 it is
Thank you
Is there a book to understand these concepts in depth
Grinstead and Snell’s Introduction to Probability, you can find it available in PDF
well done thanks a lot..
nice!! thanks
Could you tell me the book name?
It was understandable 😘👌🏻👌🏻👌🏻👌🏻
what is the name of the textbook you are using
Business Statistics by Keller. One of the best introductory stats textbooks around :)
does anyone know how to do this on the TI-84? Is it a special program that has to be installed?
you missed lambda when p(x>0)
If you want to change 6 hour to minute you multiply by 60 not divided
Hi Abdullah, thanks for your comment. I wasn't converting 6 hours to minutes. I was converting the number of customers we see per hour, to minutes. Which means, if we see 6 customers per hour, how many do we see in one minute? Do we see "6 x 60" customers per minute? Or do we see "6 / 60" customers per minute?
Hint: It is the second one. We should be seeing LESS customers per minute.
Thank you I understand
Yeah, Abdul that very true. He did the wrong thing and squad may follow so blindly
Thanks for the video and I really want to say that the way you write "lambda" is the same character of human in Chinese, ahaha
haha yea, i literally read "human" every time he writes "lambda"
How to use log?
there is an error in part B beacuse it says MORE than 10 minutes.
this means finding the propability that the service will take 10 minutes similar to done in part A (said fewer than) and then taking the answer away from 1. had the question said less then it would be correct.
you made a mistake that is P(x>5), not P(x
That's not a mistake. P(5
@@dtwtan It should be the p(x< or equal to 5) not p(x
@@johnlynch9940 Thanks John, good point. This is important if we were dealing with a discrete variable, so p(X
How could it be less probable to leave after 10 minutes than to leave before 5 minutes?
Look at the graph for Exp distribution variable, you will see that the highest probabilities are centered around 0 to the EX time taken. You can find the theoretical explanation on google.
helpful
Anyone else having a seizure?
made easy he said....
It was very good however, I don't understand how you got the answer because you did it in your head, :-(
Only a minute and a half in and I gave the like button a beat down
how did i get here
good ::D :D
wtf, this is so easy. Why do they make it so hard in class? lolololol
It was good but you did a mistake, instead of P(x
If I understand you correctly , you are saying that he should be subtracting the interval (5, infinity) from the interval (0, 8)... isn't the result of doing that subtraction the interval (0, 5) intersect (8, infinity) which would indicate the number of customers attended to within the first 5 minutes and from 8 minutes till infinity - thus a deviation from the question?
Sorry mate, I don't understand your question?
Basma, This is actually inaccurate. If he did as you described, he'd have found P(x
great lecture but you move too much
..... Save the planet!!!
You use a lot of papers
:(
Great content, terrible production quality