Hi;
I have this data: price £10<x<£40 £40<x<£50 £50<x<£60 £60<x<£80 £80<x<£200
frequency 147 109 182 317 175 total = 930.
and I want to know what percentage of items cost more than £100, Don't know where to start.
Thanks.
Hi;
I have this data: price £10<x<£40 £40<x<£50 £50<x<£60 £60<x<£80 £80<x<£200
frequency 147 109 182 317 175 total = 930.
and I want to know what percentage of items cost more than £100, Don't know where to start.
Thanks.
Are you assuming a linear distribution across each range? Because if you want over £100, then you are going to want a subset of the 175 that are between £80 and £200 which are also over £100. If you can assume that $\dfrac{200-100}{200-80} = \dfrac{5}{6}$ of the data points are between £100 and £200, then out of the total 930 data points, you can estimate that $\dfrac{\tfrac{5}{6}\cdot 175}{930} \approx 15.68\%$ of the data points are greater than £100.
80 to 200 is a range that spans 120 numbers (200-80=120). Numbers greater than 100 would all be within that range, but not every value is necessarily 100+. So, how do we distribute the values within that range? That was my question to you. I am not saying that my interpretation is correct, I am just saying that is one way to look at it.
If you have 930 data points between 10 and 200, and of those, 175 data points are between 80 and 200, how many of them are greater than 100? We have no idea. The best we can say is that no more than 175 data points are greater than 100. It is possible that no data values are greater than 100. It is also possible that all 175 data points are greater than 100. So, out of the 930 total data points, the probability of finding one that is over 100 is somewhere between $\dfrac{0}{930}$ and $\dfrac{175}{930}$.
So, what I did was assume that the data points are evenly distributed in the 80 to 200 range (that means, if we split the range into 6 parts of 20 numbers each, I am assuming that each part has an equal number of data points. So, of those 175 data points between 80 and 200, I expect about 29 between 80 and 100, another 29 between 100 and 120, another 29 between 120 and 140, another 29 between 140 and 160, etc. So, how many are more than 100? If I am correct with the distribution, there are $\dfrac{5}{6}\cdot 175$ data points that are greater than 100.
I am not sure how else to explain this. This is about as detailed as I can think to get.