I reformatted the data, so you may want to check and Edit to line up the values again. In any case, I would make the calculations based on known values for real estate that have statuses of U, S, F and C.
Instead of using dates, I would use the number of days on average between each status so that to estimate each value for a given inventory, the date can be used against the estimated number of days to forecast the date expected for the statuses of the new property.
A better and more statistically robust way would be to use external factors and run multiple regression analysis to create a prediction formula for forecasting the number of days and/or cash flow etc. from many known observations.
