Shift Alternative Consult

[gabrieltagleh]

Hi Everyone! I use xarray for financial planning models.

I´m having trouble to calculate values in arrays, where one value depends on previous period value.

Shift() function does not attends when the value of an array is calculated in function of a previous value from the same array.

I would be really thankfull if you or someone can take a I look to this case.

import xarray as xr
import pandas as pd

region_ind = pd.Index(["North","South","West"])
product_ind = pd.Index(["Orange","Apple","Banana"])

region = xr.Coordinate( 'Regions', region_ind)
product = xr.Coordinate( 'Products', product_ind)

fruits = xr.DataArray([[15.0,20.0,25.0], [6.0,5.0,0.0], [0.0,0.0,25.0]],[region, product])

periods_ind = pd.Index(["Period 1", "Period 2", "Period 3"])
periods =  xr.Coordinate( 'Periods', periods_ind)

prices =  xr.DataArray([[1,2,3],[2,4,6],[3,6,9]],[product, periods])

revenue = fruits*prices

I need to calculate the adjusted revenue of future periods as a function of a previous period value. Where I have the 99, I would like to complete with "2 Previous Period Value * X%".

flag_adjusted_revenue = xr.DataArray([1,1,0],[periods])
revenue_by_periods = xr.where(flag_adjusted_revenue==1, revenue, 99)

revenue_by_periods

[dcherian]

Hello that example code (first block) does not run. So it is hard for us to help you. In general, you should not have to use pd.Index or xr.Coordinate. See https://xarray.pydata.org/en/stable/user-guide/data-structures.html#creating-a-dataarray for how to create a dataarray directly.

[gabrieltagleh]

Hello Deepak, thank you very much for your answer. I repost the code again with the changes:

import xarray as xr
import pandas as pd

regions = ["North","South","West"]
products = ["Orange","Apple","Banana"]
q_data = [[15.0,20.0,25.0], [6.0,5.0,0.0], [0.0,0.0,25.0]]

quantities = xr.DataArray(q_data, coords=[regions, products], dims=['Regions', 'Products'])

periods = ["Period 1", "Period 2", "Period 3"]
p_data = [[1,2,3],[2,4,6],[3,6,9]]

prices =  xr.DataArray(p_data, coords=[products, periods], dims=['Products', 'Periods'])

revenue = quantities * prices

I need to calculate the adjusted revenue of future periods as a function of a previous period value. Where I have the 99, I would like to complete with "2 Previous Period Value * X%".

flag_adjusted_revenue = xr.DataArray([1,1,0], coords=[periods], dims=['Periods'])
revenue_by_periods = xr.where(flag_adjusted_revenue == 1, revenue, 99.)

revenue_by_periods

Thanks!!

[krisjune]

Have you tried the below?
revenue_by_periods = revenue
revenue_by_periods = revenue_by_periods.where(flag_adjusted_revenue == 1,revenue.shift(Periods=2)*1.15)