For truncated sample, as shown in Example 10.7, we omit the part/values which are truncated and scale up probabilities for other values. Why? Can someone explain?
Because you have no information about part of the range of possible values. And one can use method of maximaum likelihood only if the likelihood can be written as a function of parameter(s). So, if you have to express the function as likelihood, for truncated data we need to omit the value for which we have no information about parameter.