Let us suppose that we have frequencies against class interval then how we can fit a continuous distribution (gamma, normal etc) for such kind of data (when we don't have actual observation)...?
Chi-square test also applicable here, you can find the expected frequency for all intervals by using formula n*(F(U_i) - F(L_i)), Where U_i and L_i are upper and lower limit of the ith interval, respectively, and F(.) denote the distribution function.
Chi-square test also applicable here, you can find the expected frequency for all intervals by using formula n*(F(U_i) - F(L_i)), Where U_i and L_i are upper and lower limit of the ith interval, respectively, and F(.) denote the distribution function.
actually arun i want to find the estimate of the parameters for a specified distribution for such kind of data i.e. when we have not actual observation even we have data in form of frequecy.
You may found the estimates following http://www.tandfonline.com/doi/full/10.1080/03610918.2015.1076469
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software