dkgaraujo / brazilianbanks Goto Github PK

Banco Central do Brasil's IF.Data originally presents bank income statement accounts cumulatively in each semester: the values for the second and fourth quarters also contain the income and expenses of the first and third quarters, respectively.

Ideally brazilianbanks should offer users a quarterly income statement, whereby the values for each quarter only represent amounts earned or expensed during that period, with the option to retain the income data on a semester basis as originally reported.

Hi, thank you so much for creating this package. I am currently trying to use it to collect financial data for Brazil and I have encountered a few errors along the way, and I would appreciate it if you provide any idea as to overcoming them.

One of the errors I have encountered was "Error in parse_con(txt, bigint_as_char) : parse error: premature EOF" when the function is downloading the data. I tried to delete the json file and redownload it again using this function but it doesn't seem to work.
.
This error occurred when I called bank_df <- get_bank_stats(yyyymm_start = 200203, yyyymm_end = 200212)

Another error I encountered was "Error in dplyr::filter(., Report_column %in% c(5, 79, 91, 94, 98) & td == :
object 'all_data_info' not found".

This happened when I excluded the year and month of the problematic download above (I used bank_df <- get_bank_stats(yyyymm_start = 200203, yyyymm_end = 200209)), but this error will then appear instead.

I would really appreciate if you could provide any insights on this! Thank you!

As reported by a user behind a corporate firewall:

Note:

• Other uses behind the same firewall report successfully using this function

Assuming the result of a get_bank_stats() is assigned to variable "bank_data", the following code demonstrates that there are no existing record of Itaú's lending to the States in Southern Brazil for the quarter ending on June 2014:

bank_data %>% filter(FinInst == 1000080099) %>% select(Quarter, South)

This is likely to be a limited case, but still, worth checking.

Since launch of brazilianbanks, DuckDB has made impressive gains in performance speed and it supports JSON. It would be good to explore how it can be leveraged to lead to faster data loading times for users.

This issue is leading Caixa's data points that come from its financial conglomerate as NA for the quarter of 2021Q2. I haven't yet check extensively but I believe the issue does not extend to other quarters.