What is the advantage of using @Output property = new EventEmitter(); decorator with the event emitter instance when the same functionality can be achieved with @Input property: () => void

and then can pass the method from the parent class,

<selector [property]="parentMethod()">

Also, while doing output binding when @Output decorator is used,

<selector (property)="parentMethod()">

Does it internally subscribe to property event emitter instance? This only seems to be beneficial when Angular is using the same instance of a component class at multiple places (does this happen? ), couldn't think of any other reason.

Would like to understand the work which Angular is doing behind the scene and which of the above-mentioned approach is better and why?