There is an earlier video titled Lorenz Curve in Excel. That video does not include the Gini Index. It focuses on how to construct a Lorenz curve from raw data in Excel.
Thank you for this video. I took an economic class that gave us an assignment without guidance on how to create a Lorenz curve and Gini coefficient. My book only reveals formula without application. Luckily, I was able to learn from you and your videos better than the paid textbook. Thank you.
Wow, honestly THANK for this explanation. You are the first professor that I experience who explains where the Gini-coefficient originally comes from. Found the article online. Thank you!
You are very welcome. It was fun to produce all those videos, but there is no logical order to them. I just produced them as problems would come up. Now that I am retired, I am considering producing a new set of videos that follow the typical outline of a macro or micro principles class. Or, maybe a better idea would be to follow the outline of an interrmediate micro or macro class. I am still trying to figure out how best to do this.
I have searched for some tutorials to calculate GINI, and this is definitely the easiest method I found. Thank you very much for sharing the knowledge. Regards.
It is time to redefine studying by transition from traditional reading of the printed texts to practical tangible video lessons and that's the point you achieved. Thanks a lot.
Thank you. You may find it strange that I do not think the Lorenz Curve and Gini Index are very good measures of income inequality. Mathematically, there is beauty in their construction. But, they cannot measure the dynamic changes that happen to a person's income throughout their life. For example, professional people earn little income early in their careers. In those years they are in school or in apprenticeships. The Lorenz Curve would mark them poor. It would show lots of inequality. Later, as those young professionals mature and gain business experience, their incomes will rise. Many will be in top income brackets. Then the Lorenz Curve would show them rich. Again it would indicate income inequality. Of course the poor young traine is the same person as the older sucessful business professional. What is needed is a measure of income inequality across a workers lifetime.
Giovani Valdrighi Thank you for the greeting from Brazil. Over the years I have had a several students from there. They were all great students with interesting stories about home life. I have been surprised at how much interest there is about the Lorenz Curve and Gini Index. You may be surprised to know that I think they are very poor measures of true inequality. Because they are static measures, they are insensitive to the dynamic changes happening through time. For example, consider students. They measure as very poor while in school and not working, but 10 years later, after graduation and with some work experience, their income will be much higher. There are better measures such as the Instability Index. No one asks me about that. Good luck with your project. Arnold Hite
Thank you for the fantastic explanation. Would you explain further, does household number affect the population percentage? If for example, the richest individual has the highest income, but also big household number, does it affect the population percentage. And if this is the case, how do we represent the correlation between income and household number in Gini Index and Lauren Curve?
Hello Faten, I am not sure I know what you mean by household number. Do you mean the number of people in the household? For example a very rich person may have the whole extended family living together in a large compound. While a person of more modest income may live alone. Is this what you mean by household number? In the U.S. that kind of data is not available. If you have that data it would be easy to calculate a Lornz Curve and Gini Index. It would affect the population percentage and the income percentage. It would be fun to compare the two techniques.
@@arnoldhite You guess it right. Household number is the number of people in the household. if I happen to have the data for the number of people in the household, how do I go about to calculate a Lorenz curve and Gini Index? Normally, when we collect the data, we would ask how many person in the same household is having a job, and how many depends on other member(s) of the same household for living. If I have such data, should I divide the total household income with the number of people who live in the same household first before proceeding with the calculation process that you showed in this video?
Faten, Your idea is very interesting. I am happy to help you. We should exchange email addresses. My work email address is ahite (at) csuniv . edu. Do you have the data in a spreadsheet? If you will send me a small sample of the data, or just crease a sample data file of about 10 observations, I will figure out how to make your adjustments.
@@arnoldhite @arnoldhite Thank you for sharing with me your email address. I have sent you the dataset in the email and would love for you to see it. I thank you so much for kind assistance.
Sounds like a good idea. I'm kind of busy with class preps. I may do it later. But, do you have a good data set that would fit the Theil formular? If so I could use it to create the index. Also, would be fun to study the index's strengths and weaknesses. As you probably know, the Lorenz Curve is a very poor measure on inequality.
i think so yes. for example the UK's income distribution in deciles. I have found though that the entropy measures however work less well for many real world choppy distributions when you have only quartiles or deciles. The GINI is usually more robust across data sets.
Angelica Marie Gratuito I am unable to access your comment. It appears as a notification on my side of TH-cam. I cannot figure out how to reply. If you had a question about Lorenz Curves or the Gini Index, please post them as comments. I am happy to respond. Alternatively, you can find me, and my email address, on the Charleston Southern University web page.
Well with all kinds of appreciation for your time and efforts, at 04:00 it seems to me that the table that over Lorenz Curve chart has wrong values. For example, to calculate the value of C5 it will be 12.50 not 0.125. The same goes for the values of columns D and E too! Thank you.
John Adams Thank you for the note. Your point has come up before. The confusion comes from the option to express the axis in decimal (0 to 1) or in percent (0 to 100). It has been a while since I looked at that video but I bet I used percent while you used decimal. Either way the Lorenz Curve will look the same and the Gini Index will be the same. Let me know if you still feel I have an error.
Hi, I used your video to calculate the Gini Index, but I can't find the formula you have used. Would you be so kind to tell me which one you have used? A link would be enough :) Thanks a lot for making this video, it was very helpful for me. Julius
I am sorry for the confusion. The Gini Index is calculated as the ratio of two areas. First, find the area between the line of equality and the Lorenz Curve. Second, find the area on the triangle made by the line of equality and the horizontal axis. The Gini Index is the first area divided by the second. There is a short cut for calculating the second area. If you use decimals for your graph, the axis ranges from 0 to 1. That makes the second area 0.5 all the time. If your graph is in percentage (0-100%) the second area will always be 5,000. I hope this is helpful.
I followed the steps exactly, and when testing for robustness, set all of the incomes to 0, except for individual 8. However, rather than a Gini value of 1, (which should result when all the income goes to a single individual, unless I am mistaken), I got a Gini value of 0.875. What could I have done wrong?
Nick Groll, Your result sounds correct. With only 8 levels of income your Gini is correct try 100 levels of income. The Gini should be closer to one. To get a Gini of 1.00 you will need an infinite number of income classes.
So even though perfect inequality (one individual receiving all of the income) is supposed to be represented by a value of 1.00, such will not actually be shown unless there is a large enough sample size? Thanks!
Nick Groll, Yes. It sounds a little strange the way you put it but yes. Consider that when you have only eight income groups each group is 12.5% of the population. Then group 8, with all the income, is 12.5% of everyone. That is not perfect inequality. Here is another way to look at it. Let there be only two income groups. Let one group have all the income. The Gini should be 0.5. Half of the population is rich and half is poor.
Isn't the first interval from 0-20% really a triangle, and the remaining five intervals are each a combination of triangle and rectangle? I usually use this geometric approach to calculate Gini coefficients, but will have to try your version to see if it produces the same result.
Yes. Thanks for the note. Your approach yields the same result. I use to do it that way. I switched when I began using Excel. Finding the mean at the top of each rectangle seems easier to program to me.
I also think the professors have abandoned this type of inequality measures for yet another one.. I wrote a graduation thesis in 1990 on several of them...and statistical data from the USA.. Very complicated science..
Gunnar Jensen I agree that the Lorenz Curve and corresponding Gini Index are very poor measures of inequality. To me the biggest problem is the inability to track individuals as they more through different income groups. For example students typically fall in the poorest income group while students, but their futures are bright. They are not poor in the same way that others are poor their whole lives Because of data limitations it seems impossible to build an inequality measure across an individuals working life. If you know of such a measure let me know.
I spent a whole year studying this subject in 1989, reading around 300 articles, no internet then.. My thesis was 232 pages long, kept in a bank volt. Have not studied it for many years...Have been reading on economic growth and some finance since then..
Robert Shiller at Yale and Dale W. Jorgenson are very good. But there are many scholars very adept in this field..maybe too complicated with their infinite iintegrals or PDF-curves ?
Gunnar Jensen Sorry it has taken me.awhile to respond to your last posting. I have been traveling. I haven't done such a thorough search of the literature as you, but I have searched for alternative measures of income inequality. It seems to me that there are very few. There are none that attempt to measure the inequality across lifetimes. If you know of some, please let me know. I think this approach is important as the workforce becomes increasingly specialized and trained. Consider that 50 years ago workers typically went to work in a factory out of highschool. But now it is typical to study and train for several years after highschool. In those training years they will be counted as poor. Later if they have sucessful careers they will be counted as rich. Almost everyone who quotes inequality data has no idea how poor the measure is.
Jennifer, I tried hard to explain this problem in the video you watched. The explanation starts when the rectangles first appear in my video. I don't think I can explain it any better. Sorry it is not clear to you.
Hi, Thank you for sharing your knowledge. I calculate all the variable as you said in the previous video and I draw the Lorenz Curves. Everything went good, but I have some problem with calculation of the area under Lorenz curve and its sum. I verified so many times and there is no problem with functions. instead of .5, I tried with minus 5000 because my axes are100, so the area is 1000 and each part is 5000. I thing you didn't multiply in 100 for having axes of 1. Could you please explain how give a different weight to income variable, please?
I am sorry you are having a little trouble. I'm willing to help you. Can you send your data to me? I have an email address at Charleston Southern University.
polar dark I will try to help you. What problem do you have? You are welcome to contact me via email. In the case of Azartash Imani, he had two Lorenz curves. One he had scaled from 0-1. The other he scaled from 0-100. In my video the Lorenz curve is scaled from 0-1. So to find the Gini Index one must divide by 0.5. If you have a scale from 0-100 you must divide by 5,000. If you are uncertain how I got the numbers 0.5 and 5,000 send me an email at my university email address. Good luck in your studies
i have a prob with the gini index, i already try to divide it by 5,000 but the result is not make any sense, my lorenz kurve is good, but the gini index result it nearly 1
hi I was all fine until working out area A . if I use 0.5 I get a minus figure. seeing as my axis go from 0-100 would I take the area under the curve away from 50 instead?
If your scales run from 0 to 100 then the area under the triangle would be 0.5×100×100=5000. I'm not sure that answers your question. You may contact me through the Charleston Southern University website.
I am sorry for the late reply. I just saw your question. Look at the data in Excel. If you want your Lorenz Curve and Gini Index to be represented in decimal, make sure the percent of income and the percent of households are represented in decimal in Excel. For example, if households are divided into 5 groups, they could be represented as 20%, 40%, 60% 80% and 100%. This would generate a Lorenz and Gini Index in %. If you wish your Lorenz Curve and Ginin Index to be in decimal, use .20, .40, .06, .80, and 1.00.
Gunnar Jensen You are correct. Most people never consider how the Lorenz Curve must look to represent real data. Most people just assume a continuous function. I think this is true of many economic models.
please, I want to measure the extend of wages/salaries among formal public and private organizations. The survey data on wages and salaries was in range, is it possible to calculate Gini coefficient on excel with range income data. please if you have the solution, reply to help me out. Thanks
I think I can help you. Arrange your data in Excel and send it to me. I don't want to publish my email address here. You can find it on the Charleston Southern University web page.
Hello, for the % outcomes as well as the other decimal answers such as the area a/ gini/ under the lorenz how can i format my excel to round them to the same amount of digits as you have in your sheet?
There are a few different ways to format and round numerical cells in Excel. I usually right click on the cell, or group of cells, I wish to format. Then, from the drop down menu, I select the format and number of decimal places I want. I hope that helps you.
No, it could be sorted from richest to poorest. Traditionally the Lorenz Curve is drawn below the 45 degree line. If you sort from richest to poorest the Lorenz Curve would be above the 45 degree line.
@@PeopleHealthDevelop No. You must sort. Traditionally the income groups are sorted from poorest to richest. That places the Lorenz Curve under the 45 degree line of equality. You may sort richest to poorest which will place the Lorenz Curve above the 45 degree line. If you do not sort the Lorenz Curve will be meaningless.
@@arnoldhite Dear Professor, thank you very much for your help. The video and your comments help me a lot. You are an academic life saver. Regards, Sokvisal
I have a problem. When I computed are A under the Lorenz Curve the value become negative (-), did I do it incorrect way? One more thing my GIni exercise formula is Gini=1-2B. Do you have any clue to share?
The width is found by considering how many income groups you have. For example is you have the bottom 20% the bottom 40%, 60%, 80% and 100% there are 5 groups. If your horizonal axis is in percent, each group width would be 20%. If your horizonal axis is in decimal, each group width is 0.2. If you had 20 groups each width would be 5% or 0.05. I hope that helps you.
I assume the .5 you ask about is from the calculation of the area under the 45 degree line. That line, along with the X and Y axis, define the area of the triangle needed to calculate the Gini Index. First, notice the scale used on the X and Y axix to measure the % of income and the % of population. Normally, these percentages are written in decimal. So, 100 percent would be written as 1.0. This is how I have written them. The formula for the area of the triangle defined by the 45 degree line and the X and Y axis would be [(1/2)*b*h]. In the case of the Lorenz Curve this would be [(1/2)*1*1] = 0.5. I hope this is helpful. Occasionally, Lorenz Curves are drawn where the percents are represented as 100 instead or as 1.0. In these cases, the area under the triangle would be [(1/2)*100*100] = 5,000. You might be surprised to discover that the calucalution of the Gini Index would still be the same. While the area of the triangle is now 5,000 instead of 0.5, the area of section A would also be much bigger. The area A/(A+B) will be the same no matter the scale you use to define the percentages.
Gunnar Jensen Hello, I see you attempted to ask me a question. If you still have a question about the Lorenz Curve or Gini Index, write it here. I will try to answer.
+Julia Davis Thanks for the response. The reason you do not see 100% for Total Percent Income is because all of the values after the decimal point are not given. The numbers are tuncated to make the table more readable. If you recreate my table in Excel with the formulas, and reset the number of digits after the decimal to the maximum, You will see 100%, or at least 99.999999999999%. Good luck in your studies.
hi sir, i am agribusiness student i applied this gini ratio for my thesis work and i have one doubt for doing this gini ratio what are all the data are required to do this gini ratio. my thesis work farmers income on different farming systems. so, kindly clarify my doubt and it may be useful to prepare questionnaire as well for my analysis part
mani kandan Hello. I am not sure I can help you. I am wondering what you are placing on the x and y axis. Normally it would be percent of farmers and percent of farm income. Perhaps if you gave me more specific information I could help you.
Hello Justice, I'm sorry you feel I go too fast in the video. I always feel like the viewer is getting board so I try to go a little faster than normal. Perhaps if you ask me about the parts of the construction that are confusing, I can help you directly.
Sincèrement merci, j'ai beaucoup apprécié votre vidéo. Toujous sur le sujet de la méthode de calcul pour le coefficient de gini. Si vous pouviez me dire ce que vous en pensez : th-cam.com/video/7v4jHU8D0iA/w-d-xo.html Bonne continuation et à bientôt !
![[#SHORTS CLIP] ใช้ของของตัวเองไปเลยสิคะ l ซานิเบาได้เบา l One Playground](http://i.ytimg.com/vi/Ea_mTywx2p4/mqdefault.jpg)
Thank you for this video. I took an economic class that gave us an assignment without guidance on how to create a Lorenz curve and Gini coefficient. My book only reveals formula without application. Luckily, I was able to learn from you and your videos better than the paid textbook. Thank you.
You are very welcome. I had the same problem years ago. It took me awhile to work it all out. Thankfully now we have youtube.
Wow, honestly THANK for this explanation.
You are the first professor that I experience who explains where the Gini-coefficient originally comes from. Found the article online.
Thank you!
You are very welcome. It was fun to produce all those videos, but there is no logical order to them. I just produced them as problems would come up. Now that I am retired, I am considering producing a new set of videos that follow the typical outline of a macro or micro principles class. Or, maybe a better idea would be to follow the outline of an interrmediate micro or macro class. I am still trying to figure out how best to do this.
I have searched for some tutorials to calculate GINI, and this is definitely the easiest method I found. Thank you very much for sharing the knowledge. Regards.
It is time to redefine studying by transition from traditional reading of the printed texts to practical tangible video lessons and that's the point you achieved. Thanks a lot.
thanks you so much Arnold for this amazing video! you have no idea how much this helped me. god bless you 🙏🏻
Thanks a lot sir, it's even helping me after 9yrs of upload
You are very welcome. I am retired now. I often wonder if I should update all those old videos. Maybe I should produce new ones on different topics.
i have almost forgot that but your video help me to understand the Gini coefficient. thank you sir
Thank you so much! This is helpful. No one explains this like in this video. Best Regards.
Excellent Arnold!! my students are so happy, because your explanation es deep and nice!!
Thank you. You may find it strange that I do not think the Lorenz Curve and Gini Index are very good measures of income inequality. Mathematically, there is beauty in their construction. But, they cannot measure the dynamic changes that happen to a person's income throughout their life. For example, professional people earn little income early in their careers. In those years they are in school or in apprenticeships. The Lorenz Curve would mark them poor. It would show lots of inequality. Later, as those young professionals mature and gain business experience, their incomes will rise. Many will be in top income brackets. Then the Lorenz Curve would show them rich. Again it would indicate income inequality. Of course the poor young traine is the same person as the older sucessful business professional. What is needed is a measure of income inequality across a workers lifetime.
Thank you so much! You were a lifesaver for my economics class.
Thank you. This is super helpful. I have always struggled with this calculation. Hopefully, no more
Thank You Sir, very simple presentation and legible video to comprehend Lorenz curve and Gini coefficient
Your simple Excel setup looks great I would like to experiment with such a program..
Hi, thanks for de video!
It really helped me
Greetings from Brazil
Giovani Valdrighi Thank you for the greeting from Brazil. Over the years I have had a several students from there. They were all great students with interesting stories about home life.
I have been surprised at how much interest there is about the Lorenz Curve and Gini Index. You may be surprised to know that I think they are very poor measures of true inequality. Because they are static measures, they are insensitive to the dynamic changes happening through time. For example, consider students. They measure as very poor while in school and not working, but 10 years later, after graduation and with some work experience, their income will be much higher.
There are better measures such as the Instability Index. No one asks me about that. Good luck with your project.
Arnold Hite
thank you, the video was very helpful for my exercise.
Thank you so much, very helpful and easy to understand
Thank you arnoldhite! You make my Gini coefficient calculus easier.
Thank you for the fantastic explanation. Would you explain further, does household number affect the population percentage? If for example, the richest individual has the highest income, but also big household number, does it affect the population percentage. And if this is the case, how do we represent the correlation between income and household number in Gini Index and Lauren Curve?
Hello Faten, I am not sure I know what you mean by household number. Do you mean the number of people in the household? For example a very rich person may have the whole extended family living together in a large compound. While a person of more modest income may live alone. Is this what you mean by household number? In the U.S. that kind of data is not available. If you have that data it would be easy to calculate a Lornz Curve and Gini Index. It would affect the population percentage and the income percentage. It would be fun to compare the two techniques.
@@arnoldhite You guess it right. Household number is the number of people in the household. if I happen to have the data for the number of people in the household, how do I go about to calculate a Lorenz curve and Gini Index? Normally, when we collect the data, we would ask how many person in the same household is having a job, and how many depends on other member(s) of the same household for living. If I have such data, should I divide the total household income with the number of people who live in the same household first before proceeding with the calculation process that you showed in this video?
Faten, Your idea is very interesting. I am happy to help you. We should exchange email addresses. My work email address is ahite (at) csuniv . edu. Do you have the data in a spreadsheet? If you will send me a small sample of the data, or just crease a sample data file of about 10 observations, I will figure out how to make your adjustments.
@@arnoldhite @arnoldhite Thank you for sharing with me your email address. I have sent you the dataset in the email and would love for you to see it. I thank you so much for kind assistance.
Thanks....you made my work easier
Hi Arnold
Thank you so much for your help.
You have helped me. Thank you.
Great vid. Could you also explain how to construct other Inequality measures from this such as Theil Index and more?
Sounds like a good idea. I'm kind of busy with class preps. I may do it later. But, do you have a good data set that would fit the Theil formular? If so I could use it to create the index. Also, would be fun to study the index's strengths and weaknesses. As you probably know, the Lorenz Curve is a very poor measure on inequality.
i think so yes. for example the UK's income distribution in deciles. I have found though that the entropy measures however work less well for many real world choppy distributions when you have only quartiles or deciles. The GINI is usually more robust across data sets.
Thanks a lot teacher. I appreciate forever. Take care.
Angelica Marie Gratuito
I am unable to access your comment. It appears as a notification on my side of TH-cam. I cannot figure out how to reply. If you had a question about Lorenz Curves or the Gini Index, please post them as comments. I am happy to respond. Alternatively, you can find me, and my email address, on the Charleston Southern University web page.
Well with all kinds of appreciation for your time and efforts, at 04:00 it seems to me that the table that over Lorenz Curve chart has wrong values. For example, to calculate the value of C5 it will be 12.50 not 0.125. The same goes for the values of columns D and E too! Thank you.
John Adams Thank you for the note. Your point has come up before. The confusion comes from the option to express the axis in decimal (0 to 1) or in percent (0 to 100). It has been a while since I looked at that video but I bet I used percent while you used decimal. Either way the Lorenz Curve will look the same and the Gini Index will be the same. Let me know if you still feel I have an error.
Great explanation, thanks dude!!!
Really helpful
Exactly what I was looking for.
Amazing explanation
Thank you very much, Sir.
Many Thanks, it is very useful..
Great presentation
It is great help!
excellent video!
Very useful thanks!
Hi, I used your video to calculate the Gini Index, but I can't find the formula you have used. Would you be so kind to tell me which one you have used? A link would be enough :)
Thanks a lot for making this video, it was very helpful for me.
Julius
I am sorry for the confusion. The Gini Index is calculated as the ratio of two areas. First, find the area between the line of equality and the Lorenz Curve. Second, find the area on the triangle made by the line of equality and the horizontal axis. The Gini Index is the first area divided by the second.
There is a short cut for calculating the second area. If you use decimals for your graph, the axis ranges from 0 to 1. That makes the second area 0.5 all the time. If your graph is in percentage (0-100%) the second area will always be 5,000. I hope this is helpful.
Thank you so much!
thank you for your help!
crystal clear... thanks.
thanks a lot; nice lecture
Fantastic!
I followed the steps exactly, and when testing for robustness, set all of the incomes to 0, except for individual 8. However, rather than a Gini value of 1, (which should result when all the income goes to a single individual, unless I am mistaken), I got a Gini value of 0.875. What could I have done wrong?
Nick Groll, Your result sounds correct. With only 8 levels of income your Gini is correct
try 100 levels of income. The Gini should be closer to one. To get a Gini of 1.00 you will need an infinite number of income classes.
So even though perfect inequality (one individual receiving all of the income) is supposed to be represented by a value of 1.00, such will not actually be shown unless there is a large enough sample size? Thanks!
Nick Groll, Yes. It sounds a little strange the way you put it but yes. Consider that when you have only eight income groups each group is 12.5% of the population. Then group 8, with all the income, is 12.5% of everyone. That is not perfect inequality. Here is another way to look at it. Let there be only two income groups. Let one group have all the income. The Gini should be 0.5. Half of the population is rich and half is poor.
That helps, perfect! Thanks again!
You can normalize the coefficient by multiplying it by (n/n-1) and it will make sense.
you are my hero!
thanks👍
Isn't the first interval from 0-20% really a triangle, and the remaining five intervals are each a combination of triangle and rectangle? I usually use this geometric approach to calculate Gini coefficients, but will have to try your version to see if it produces the same result.
Yes. Thanks for the note. Your approach yields the same result. I use to do it that way. I switched when I began using Excel. Finding the mean at the top of each rectangle seems easier to program to me.
@@arnoldhite Ah, good to know. I will definitely give this a try.
Hi mr Hite. What about cases with negative income or values? Do you have a good approach for that?
I also think the professors have abandoned this type of inequality measures for yet another one.. I wrote a graduation thesis in 1990 on several of them...and statistical data from the USA.. Very complicated science..
Gunnar Jensen
I agree that the Lorenz Curve and corresponding Gini Index are very poor measures of inequality. To me the biggest problem is the inability to track individuals as they more through different income groups. For example students typically fall in the poorest income group while students, but their futures are bright. They are not poor in the same way that others are poor their whole lives
Because of data limitations it seems impossible to build an inequality measure across an individuals working life.
If you know of such a measure let me know.
I spent a whole year studying this subject in 1989, reading around 300 articles, no internet then.. My thesis was 232 pages long, kept in a bank volt. Have not studied it for many years...Have been reading on economic growth and some finance since then..
Robert Shiller at Yale and Dale W. Jorgenson are very good. But there are many scholars very adept in this field..maybe too complicated with their infinite iintegrals or PDF-curves ?
Anyone can practice with your setup..
Gunnar Jensen
Sorry it has taken me.awhile to respond to your last posting. I have been traveling. I haven't done such a thorough search of the literature as you, but I have searched for alternative measures of income inequality. It seems to me that there are very few. There are none that attempt to measure the inequality across lifetimes. If you know of some, please let me know.
I think this approach is important as the workforce becomes increasingly specialized and trained. Consider that 50 years ago workers typically went to work in a factory out of highschool. But now it is typical to study and train for several years after highschool. In those training years they will be counted as poor. Later if they have sucessful careers they will be counted as rich.
Almost everyone who quotes inequality data has no idea how poor the measure is.
How do you know that both ways of trying to find the area using rectangles overestimate and underestimate area A?
Jennifer,
I tried hard to explain this problem in the video you watched. The explanation starts when the rectangles first appear in my video. I don't think I can explain it any better. Sorry it is not clear to you.
Hi, Thank you for sharing your knowledge. I calculate all the variable as you said in the previous video and I draw the Lorenz Curves. Everything went good, but I have some problem with calculation of the area under Lorenz curve and its sum. I verified so many times and there is no problem with functions. instead of .5, I tried with minus 5000 because my axes are100, so the area is 1000 and each part is 5000. I thing you didn't multiply in 100 for having axes of 1. Could you please explain how give a different weight to income variable, please?
I am sorry you are having a little trouble. I'm willing to help you. Can you send your data to me? I have an email address at Charleston Southern University.
i have a same problem sir
polar dark
I will try to help you. What problem do you have? You are welcome to contact me via email. In the case of Azartash Imani, he had two Lorenz curves. One he had scaled from 0-1. The other he scaled from 0-100. In my video the Lorenz curve is scaled from 0-1. So to find the Gini Index one must divide by 0.5. If you have a scale from 0-100 you must divide by 5,000. If you are uncertain how I got the numbers 0.5 and 5,000 send me an email at my university email address. Good luck in your studies
i have a prob with the gini index, i already try to divide it by 5,000 but the result is not make any sense, my lorenz kurve is good, but the gini index result it nearly 1
You are welcome to send me your data. I will help you find your error. I have an email address at Charleston Southern University.
Thank you!!!!
Thank you very much)))
Tx sir
Mr Arnoldhite , This application which theorem ? I search İncome equality says a lot of theorem . Could you help me pls
SIR, thank you
hi I was all fine until working out area A . if I use 0.5 I get a minus figure. seeing as my axis go from 0-100 would I take the area under the curve away from 50 instead?
If your scales run from 0 to 100 then the area under the triangle would be
0.5×100×100=5000. I'm not sure that answers your question. You may contact me through the Charleston Southern University website.
sir how did you change the data number to decimals less than 1??
I am sorry for the late reply. I just saw your question. Look at the data in Excel. If you want your Lorenz Curve and Gini Index to be represented in decimal, make sure the percent of income and the percent of households are represented in decimal in Excel. For example, if households are divided into 5 groups, they could be represented as 20%, 40%, 60% 80% and 100%. This would generate a Lorenz and Gini Index in %. If you wish your Lorenz Curve and Ginin Index to be in decimal, use .20, .40, .06, .80, and 1.00.
You are using a simple excel program ?
Gunnar Jensen
Yes. I used MS Office.
Amazing. The conventional Gini formula does not look so simple...or discrete.. Looks more like infinitely continuous..
Gunnar Jensen
You are correct. Most people never consider how the Lorenz Curve must look to represent real data. Most people just assume a continuous function. I think this is true of many economic models.
please, I want to measure the extend of wages/salaries among formal public and private organizations. The survey data on wages and salaries was in range, is it possible to calculate Gini coefficient on excel with range income data. please if you have the solution, reply to help me out. Thanks
I think I can help you. Arrange your data in Excel and send it to me. I don't want to publish my email address here. You can find it on the Charleston Southern University web page.
Hello, for the % outcomes as well as the other decimal answers such as the area a/ gini/ under the lorenz how can i format my excel to round them to the same amount of digits as you have in your sheet?
There are a few different ways to format and round numerical cells in Excel. I usually right click on the cell, or group of cells, I wish to format. Then, from the drop down menu, I select the format and number of decimal places I want. I hope that helps you.
@@arnoldhite Thank you so much for the quick reply!
Extremely complicated econometrics estimations with assumptions of normally distributed population attributes..
what were the variable data given here BCDE I dont really get them
Sorry, I do not understand your question. Ask me a specific question. I will try to help you.
Please tell me how you get the % population column?
Does it alway need to sort from lowest to the richest before calculate Lorenz Curve?
No, it could be sorted from richest to poorest. Traditionally the Lorenz Curve is drawn below the 45 degree line. If you sort from richest to poorest the Lorenz Curve would be above the 45 degree line.
@@arnoldhite Thank you for your respond. So, it is ok, if I construct Lorenz Curve without sorting?
@@PeopleHealthDevelop No. You must sort. Traditionally the income groups are sorted from poorest to richest. That places the Lorenz Curve under the 45 degree line of equality. You may sort richest to poorest which will place the Lorenz Curve above the 45 degree line. If you do not sort the Lorenz Curve will be meaningless.
@@arnoldhite Dear Professor, thank you very much for your help. The video and your comments help me a lot. You are an academic life saver. Regards, Sokvisal
I have a problem. When I computed are A under the Lorenz Curve the value become negative (-), did I do it incorrect way? One more thing my GIni exercise formula is Gini=1-2B. Do you have any clue to share?
Tell how did u get the width
The width is found by considering how many income groups you have. For example is you have the bottom 20% the bottom 40%, 60%, 80% and 100% there are 5 groups. If your horizonal axis is in percent, each group width would be 20%. If your horizonal axis is in decimal, each group width is 0.2. If you had 20 groups each width would be 5% or 0.05. I hope that helps you.
.5 is from where?
I assume the .5 you ask about is from the calculation of the area under the 45 degree line. That line, along with the X and Y axis, define the area of the triangle needed to calculate the Gini Index. First, notice the scale used on the X and Y axix to measure the % of income and the % of population. Normally, these percentages are written in decimal. So, 100 percent would be written as 1.0. This is how I have written them. The formula for the area of the triangle defined by the 45 degree line and the X and Y axis would be [(1/2)*b*h]. In the case of the Lorenz Curve this would be [(1/2)*1*1] = 0.5. I hope this is helpful.
Occasionally, Lorenz Curves are drawn where the percents are represented as 100 instead or as 1.0. In these cases, the area under the triangle would be [(1/2)*100*100] = 5,000. You might be surprised to discover that the calucalution of the Gini Index would still be the same. While the area of the triangle is now 5,000 instead of 0.5, the area of section A would also be much bigger. The area A/(A+B) will be the same no matter the scale you use to define the percentages.
Using Excel ?
Gunnar Jensen
Hello, I see you attempted to ask me a question. If you still have a question about the Lorenz Curve or Gini Index, write it here. I will try to answer.
I love you.
hi
there is a calculation mistake in the excel sheet
Julia,
If I have made a mistake, I want to correct it. Can you identify where I have made a mistake?
the mistake is in the calculation of the percentage of the income. If you add up all percentages you end up with 0.999 and not with 1
+Julia Davis
Thanks for the response. The reason you do not see 100% for Total Percent Income is because all of the values after the decimal point are not given. The numbers are tuncated to make the table more readable. If you recreate my table in Excel with the formulas, and reset the number of digits after the decimal to the maximum, You will see 100%, or at least 99.999999999999%. Good luck in your studies.
hi sir, i am agribusiness student i applied this gini ratio for my thesis work and i have one doubt for doing this gini ratio what are all the data are required to do this gini ratio. my thesis work farmers income on different farming systems. so, kindly clarify my doubt and it may be useful to prepare questionnaire as well for my analysis part
mani kandan
Hello. I am not sure I can help you. I am wondering what you are placing on the x and y axis. Normally it would be percent of farmers and percent of farm income. Perhaps if you gave me more specific information I could help you.
It was easy
The video is excellent but the tutor is too fast...!!!!
Hello Justice,
I'm sorry you feel I go too fast in the video. I always feel like the viewer is getting board so I try to go a little faster than normal. Perhaps if you ask me about the parts of the construction that are confusing, I can help you directly.
Sincèrement merci, j'ai beaucoup apprécié votre vidéo. Toujous sur le sujet de la méthode de calcul pour le coefficient de gini. Si vous pouviez me dire ce que vous en pensez : th-cam.com/video/7v4jHU8D0iA/w-d-xo.html Bonne continuation et à bientôt !
Thank you so much !!!
Thank You so much!