By default Causal uses calendar years (January to December) for the annual aggregation in charts/tables, however if you'd like to use fiscal years, you can do that by :
If your model has variables with uncertainties (e.g. "1 to 2" or "normal(3,1)"), Causal runs many simulations of the model, taking new samples from the distributions every time (monte carlo simulation). The round function is applied to the individual simulation results, but the average of the rounded simulations results might not always be a rounded number. For example an "input" variable with the formula "1 to 2" is a triangle distribution (left picture). If you round this variable with "round(input)", Causal will take samples from the input variable and round them. 50% of the samples will be between 1 and 1.5 and 50% of the samples will be between 1.5 and 2. So the rounded variable is a distribution where 50% of the values are 1 and 50% of the values are 2 (right picture). The average value is 1.5, even though the individual simulation results never have the value 1.5.
Causal uses Monte-Carlo simulations for uncertain values which is why refreshing the page may cause the values to change.
No, Causal doesn't support circular references for the following reasons: