Try to explain MLE to other people who dont know statistic

Of course, I can explain Maximum Likelihood Estimation and likelihood, let us

start on figure out what is likelihood first. Likelihood is a very important and common

word that you can see in statistical inference. It has similar meaning with

“probability”, but in statistic, they were completely different. For example, let us say

we had a model which made by a sample of data with values of the unknown

parameters, the likelihood measures the how well it fits a set of observation of the

model. Compare to the probability, probability is the value of part of area which under

the distribution model. Likelihood is formed from the joint of the sample, it is the y-

axis value with the distribution model.

When you already know the difference between Likelihood and Probability, we

can start to discuss the meaning of Maximum Likelihood Estimation. In simple terms,

each parameter in distribution can make their own likelihood value, but the relation

with distribution should be high. For getting the highest likelihood value, the location

of the center of the distribution should be the highest, too. This is called the

maximizes the likelihood value of observing the parameters we measured. This can be

used to find the maximum likelihood estimation of mean, also can be used in standard

deviation. Did that explanation help? You can find the formula of calculate maximum

likelihood estimation for more detail understanding.

Reference:

Brooks-Bartlett, J. (2018, Jan 3, 2018).

Maximum likelihood estimation. Retrieved from

https://towardsdatascience.com/probability-concepts-explained-maximum-likelihood-

estimation-c7b4342fdbb1

Any opinion? Please leave your comment below