Gretl time serie mutiple records on 1 date for both historic and to be forecast data


I have a data set of time serie records with both the exogenous (numbers) and endogenous (numbers) collumns filled. The endogenous collumns are partially not real numbers but dummie collumns of 0 or 1 for different neighbourhoods (if something is is in the neighbourhood south its gets an 1 in the collumn south). Can anyone tell me which time serie model i should use (1).
Some days have no records and some days have more than 1 records. Is it possible with gretl to let gretl figure out the equotation?(2)
I already know how to do that if there is just one record per day. In the automatically generated equotation i dont see the the time variable. Can anyone tell me how to produce the equotation including the time variable? (3)

thanks in advance for your reply