I am trying to construct a bimodal normal distribution from curve parameters like mean (two for two modes) and standard deviation (two). I wanted to know what are the other parameters required for this. Also if there's any specific equation to deal with such distributions.
I have tried adding two random normal distributions in matlab and it gives me a bimodal distribution. But I am not sure whether it is the best way to do it. This is because I also want to include the fraction/weight of each mode in the distribution (have no idea how to do it).
I am quite new to statistics and would really appreciate your suggestions.
Thank you very much for your help.
Basically there are five parameters.
Two for each peak and one that affect how blend together.
You can experiment on this link
https://demonstrations.wolfram.com/BimodalNormalDistributionMixtures/
and see how each of them affect final distribution.
Also, you could look here
https://www.statisticshowto.datasciencecentral.com/what-is-a-bimodal-distribution/
and here
https://psychology.wikia.org/wiki/Bimodal_distribution
Hope that this helps
- Badges
- Science topic