downloading data from OECD website using the OECD package in R

user_13954869 0
Viewed 767

I am trying to download data directly from the OECD website using the OECD package in R. I'm specifically trying to download data from the industrial production dataset (https://data.oecd.org/industry/industrial-production.htm) for South Africa. I believe that the codes for the dataset itself and for South Africa are MEI_REAL and ZAF.

However, when I try to run the following

df <- get_dataset("MEI_REAL",
                  filter = 'ZAF',
                  start_time = 2019, end_time = 2020)

I get the following error

Error in rsdmx::readSDMX(url) : 
  HTTP request failed with status: 400 Bad Request

Can anyone advise on what I'm doing wrong? I've never used this package before so I'm struggling to figure it out.

TIA

r
3 Answers

To actually use the filter it has to be something like

df <- get_dataset("MEI_REAL",
                  filter = list(c(),'ZAF'),
                  start_time = 2019, end_time = 2020)

as the country codes are in the second column, but I don't know how you can know this in advance without just downloading the whole dataset first (just drop the filter argument entirely to do that).

user_132039710

The code in the question fails because LOCATION is the second column in the data frame, and the filter = statement does not account for this. We can fix the request by adding NULL to the list() passed to filter= argument.

library(OECD)

# filter on second column
saProduction <- get_dataset("MEI_REAL",
                            filter = list(NULL,"ZAF"),
                            start_time = 2019,
                            end_time = 2020)

head(saProduction)

...and the output:

> head(saProduction)
   SUBJECT LOCATION FREQUENCY TIME_FORMAT UNIT POWERCODE REFERENCEPERIOD obsTime
1 PRMNTO01      ZAF         A         P1Y  IDX         0        2015_100    2019
2 PRMNTO01      ZAF         Q         P3M  IDX         0        2015_100 2019-Q1
3 PRMNTO01      ZAF         Q         P3M  IDX         0        2015_100 2019-Q2
4 PRMNTO01      ZAF         Q         P3M  IDX         0        2015_100 2019-Q3
5 PRMNTO01      ZAF         Q         P3M  IDX         0        2015_100 2019-Q4
6 PRMNTO01      ZAF         Q         P3M  IDX         0        2015_100 2020-Q1
   obsValue
1 100.46670
2 100.97510
3 101.30840
4 100.04170
5  99.47495
6  97.40812

How did I figure out that the right entry into the list is NULL? When one looks at the arguments for get_dataset(), we see that NULL is a valid value for filter =, so I inferred that I could use it as a value in the filter list().

user_84719310

For anyone who ecnounters the same problem: My "solution" is to set the dataframe-ID you want at different positions in the filter. E.g: 1.try:

filter_GDP <- list("B1_GE")
GDP <- get_dataset("QNA", filter_GDP)

This gives back an error, hence i set the dataframe-ID at the next position in the filter list (by including NULL):

filter_GDP <- list(NULL, "B1_GE")
GDP <- get_dataset("QNA", filter_GDP)

This works, so now you can look up the positions of the other parameters you want to filter, in my case:

filter_GDP <- list(NULL,"B1_GE", "CQRSA", "Q")
GDP <- get_dataset("QNA", filter_GDP)
user_118012690
Related