math - Statistical estimation algorithm -
I'm not sure whether this question is suitable for stack overflow, but I'll try it in any way. I have some data as follows:
I There is another set of data that I believe is an equal distribution, but I know only the total percentage (such as 30% instead of 17%). On a new total% basis for each individual level,% and the original distribution?
The question is not clear if you add a new total percentage by adding additional data If you want to, you should have a quantity related to your percentage so that you can make an average weighted average.
If you want to determine that the new set of data has a different distribution than the historical data, then there are many examinations, which is the estimated percentage of the values falling under most typical numbers is made of. Comparing the distribution of two populations, there are a lot of literature on the subject.
There is a standard method for coupled samples if you can not make any assumptions regarding the distribution of data. Non-coupled data exists but they require some intensive study.
Comments
Post a Comment