Answer:
The Normal distribution is a continuous probability distribution with possible values all the reals. Some properties of this distribution are:
Is symmetrical and bell shaped no matter the parameters used. Usually if X is a random variable normally distributed we write this like that:

The two parameters are:
who represent the mean and is on the center of the distribution
who represent the standard deviation
One particular case is the normal standard distribution denoted by:

Example: Usually this distribution is used to model almost all the practical things in the life one of the examples is when we can model the scores of a test. Usually the distribution for this variable is normally distributed and we can find quantiles and probabilities associated
Step-by-step explanation:
The Normal distribution is a continuous probability distribution with possible values all the reals. Some properties of this distribution are:
Is symmetrical and bell shaped no matter the parameters used. Usually if X is a random variable normally distributed we write this like that:

The two parameters are:
who represent the mean and is on the center of the distribution
who represent the standard deviation
One particular case is the normal standard distribution denoted by:

Example: Usually this distribution is used to model almost all the practical things in the life one of the examples is when we can model the scores of a test. Usually the distribution for this variable is normally distributed and we can find quantiles and probabilities associated