What data structure to use for financial model

[romanpop]

Hi all,

I am new to R and I am trying to put my financial model from excel to R (sounds weird but I actually got some progress. The question is - the structure of my model is like that:

Account_1 (i.e. Revenue)
2020 2021 2022 2023 ...
obj_1 ... ...
obj_2
obj_3
... ... ...

Account_2 (i.e. Prices)
2020 2021 2022 2023 ...
obj_1 ... ...
obj_2
obj_3
... ... ...

Account_3 (i.e. Volumes)
2020 2021 2022 2023 ...
obj_1 ... ...
obj_2
obj_3
... ... ...

Obviously, Revenue = Prices * Volumes. If there was only one object, then I would turn them into named vectors and do pretty simple vector arithmetics. I could also mutate them into one dataframe with Years as an index.
But with multiple objects, I don't know how to deal with such 2-dimensional array. The goal is to simplify overall arithmetics as possible, but also to mutate these 2-dimensional different accounts in some sort of df.
I need also to be able to refer (+filter, +subset) to either object or year of such array.

What data structure would you suggest me to use? Matrix or lists or what?

Thank you,

[romanpop]

Image is always better

[arthur.t]

One data frame with columns for year, object name, volume, price, revenue.

[arthur.t]

It’s generally best to have

• one large data frame instead of many smalls
• columns that contain everything of same measure (one revenue column instead of multiple)
• long format instead of wide (years in rows not columns)

This will allow you to most easily aggregate and plot.

‘df %>% group_by(product) %>% summarize(sum(revenue))’

‘df %>% ggplot() + aes(year, revenue, color = product) + geom_line()’

[romanpop]

Thank you!

Looks like the best option and most corresponding to "tidy" approach