I want to get data using World Bank's API. For this purpose I use follow query.
wb_data <- httr::GET("http://api.worldbank.org/v2/country/all/indicator/AG.AGR.TRAC.NO?format=json") %>%
content("text", encoding = "UTF-8") %>%
fromJSON(flatten = T) %>%
data.frame()
It works pretty good. However, when I try to specify more than two variables it doesn't work.
http://api.worldbank.org/v2/country/all/indicator/AG.AGR.TRAC.NO;NE.CON.PRVT.ZS?format=json
Note, if i change format to xml and also add source=2 because data become from same database (World Development Indicator) query works.
http://api.worldbank.org/v2/country/all/indicator/AG.AGR.TRAC.NO;NE.CON.PRVT.ZS?source=2&formal=xml
However, if i want to get data from different databases (e.g. WDI and Doing Business) it doesn't work again.
So, my first question is how can I get multiple data from different databases using one query. According to the World Bank API tutorial I can include about 60 indicators.
My second question is how can I specify number of rows per page. As I might know I can add something like &per_page=100 to get 100 rows as an output. Should i calculate number of rows by myself or I can use something lika that &per_page=9999999 to get all data upon request.
P.S. I don't want to use any libraries (such as: wb or wbstats). I want to do it by myself and also to learn something new.
Here's an answer to your question. To use multiple indicators and return JSON, you need to provide both the source ID and the format type, as mentioned in the World Bank API tutorial. You can get the total number of pages from one of the returned JSON parameters, called "total". You can then use this value in a second GET request to return the full number of pages using the per_page parameter.
library(magrittr)
library(httr)
library(jsonlite)
# set up the target url - you need BOTH the source ID and the format parameters
target_url <- "http://api.worldbank.org/v2/country/chn;ago/indicator/AG.AGR.TRAC.NO;SP.POP.TOTL?source=2&format=json"
# look at the metadata returned for the target url
httr::GET(target_url) %>%
content("text", encoding = "UTF-8") %>%
fromJSON(flatten = T) %>%
# the metadata is in the first item in the returned list of JSON
extract2(1)
#> $page
#> [1] 1
#>
#> $pages
#> [1] 5
#>
#> $per_page
#> [1] 50
#>
#> $total
#> [1] 240
#>
#> $sourceid
#> NULL
#>
#> $lastupdated
#> [1] "2019-12-20"
# get the total number of pages for the target url query
wb_data_totalpagenumber <- httr::GET(target_url) %>%
content("text", encoding = "UTF-8") %>%
fromJSON(flatten = T) %>%
# get the first item in the returned list of JSON
extract2(1) %>%
# get the total number of pages, which is a named element called "total"
extract2("total")
# get the data
wb_data <- httr::GET(paste0(target_url, "&per_page=", wb_data_totalpagenumber)) %>%
content("text", encoding = "UTF-8") %>%
fromJSON(flatten = T) %>%
# get the data, which is the second item in the returned list of JSON
extract2(2) %>%
data.frame()
# look at the data
dim(wb_data)
#> [1] 240 11
head(wb_data)
#> countryiso3code date value scale unit obs_status decimal indicator.id
#> 1 AGO 2019 NA 0 AG.AGR.TRAC.NO
#> 2 AGO 2018 NA 0 AG.AGR.TRAC.NO
#> 3 AGO 2017 NA 0 AG.AGR.TRAC.NO
#> 4 AGO 2016 NA 0 AG.AGR.TRAC.NO
#> 5 AGO 2015 NA 0 AG.AGR.TRAC.NO
#> 6 AGO 2014 NA 0 AG.AGR.TRAC.NO
#> indicator.value country.id country.value
#> 1 Agricultural machinery, tractors AO Angola
#> 2 Agricultural machinery, tractors AO Angola
#> 3 Agricultural machinery, tractors AO Angola
#> 4 Agricultural machinery, tractors AO Angola
#> 5 Agricultural machinery, tractors AO Angola
#> 6 Agricultural machinery, tractors AO Angola
tail(wb_data)
#> countryiso3code date value scale unit obs_status decimal indicator.id
#> 235 CHN 1965 715185000 <NA> 0 SP.POP.TOTL
#> 236 CHN 1964 698355000 <NA> 0 SP.POP.TOTL
#> 237 CHN 1963 682335000 <NA> 0 SP.POP.TOTL
#> 238 CHN 1962 665770000 <NA> 0 SP.POP.TOTL
#> 239 CHN 1961 660330000 <NA> 0 SP.POP.TOTL
#> 240 CHN 1960 667070000 <NA> 0 SP.POP.TOTL
#> indicator.value country.id country.value
#> 235 Population, total CN China
#> 236 Population, total CN China
#> 237 Population, total CN China
#> 238 Population, total CN China
#> 239 Population, total CN China
#> 240 Population, total CN China
Created on 2020-01-30 by the reprex package (v0.3.0)
Related
I've written a function in R using pdftools to read a table from a pdf. The function gets the job done, but unfortunately the table contains a column for notes, which is only partially filled. As a result the data in the resulting table is shifted by one column in the row containing a note.
Here's the table.
And here's the code:
# load library
library(pdftools)
# link to report
url <- "https://www.rymanhealthcare.co.nz/hubfs/Investor%20Centre/Financial/Half%20year%20results%202022/Ryman%20Healthcare%20Limited%20-%20Announcement%20Numbers%20and%20financial%20statements%20-%2030%20September%202022.pdf"
# read data through pdftool
data <- pdf_text(url)
# create a function to read the pdfs
scrape_pdf <- function(list_of_tables,
table_number,
number_columns,
column_names,
first_row,
last_row) {
data <- list_of_tables[table_number]
data <- trimws(data)
data <- strsplit(data, "\n")
data <- data[[1]]
data <- data[min(grep(first_row, data)):
max(grep(last_row, data))]
data <- str_split_fixed(data, " {2,}", number_columns)
data <- data.frame(data)
names(data) <- column_names
return(data)
}
names <- c("","6m 30-9-2022","6m 30-9-2021","12m 30-3-2022")
output <- scrape_pdf(rym22Q3fs,3,5,names,"Care fees","Basic and diluted")
And the output.
6m 30-9-2022 6m 30-9-2021 12m 30-3-2022 NA
1 Care fees 210,187 194,603 398,206
2 Management fees 59,746 50,959 105,552
3 Interest received 364 42 41
4 Other income 3,942 2,260 4,998
5 Total revenue 274,239 247,864 508,797
6
7 Fair-value movement of
8 investment properties 3 261,346 285,143 745,885
9 Total income 535,585 533,007 1,254,682
10
11 Operating expenses (265,148) (225,380) (466,238)
12 Depreciation and
13 amortisation expenses (22,996) (17,854) (35,698)
14 Finance costs (19,355) (15,250) (30,664)
15 Impairment loss 2 (10,784) - -
16 Total expenses (318,283) (258,484) (532,600)
17
18 Profit before income tax 217,302 274,523 722,082
19 Income tax (expense) / credit (23,316) 6,944 (29,209)
20 Profit for the period 193,986 281,467 692,873
21
22 Earnings per share
23 Basic and diluted (cents per share) 38.8 56.3 138.6
How can I best circumvent this issue?
Many thanks in advance!
While readr::read_fwf() is for handling fixed width files, it performs pretty well on text from pdftools too once header / footer rows are removed. Even if it has to guess column widths, though those can be specified too.
library(pdftools)
library(dplyr, warn.conflicts = F)
url <- "https://www.rymanhealthcare.co.nz/hubfs/Investor%20Centre/Financial/Half%20year%20results%202022/Ryman%20Healthcare%20Limited%20-%20Announcement%20Numbers%20and%20financial%20statements%20-%2030%20September%202022.pdf"
data <- pdf_text(url)
scrape_pdf <- function(pdf_text_item, first_row_str, last_row_str){
lines <- unlist(strsplit(pdf_text_item, "\n"))
# remove 0-length lines
lines <- lines[nchar(lines) > 0]
lines <- lines[min(grep(first_row_str, lines)):
max(grep(last_row_str , lines))]
# paste lines back into single string for read_fwf()
paste(lines, collapse = "\n") %>%
readr::read_fwf() %>%
# re-connect strings in first colum if values were split between rows
mutate(X1 = if_else(!is.na(lag(X1)) & is.na(lag(X3)), paste(lag(X1), X1), X1)) %>%
filter(!is.na(X3))
}
output <- scrape_pdf(data[3], "Care fees","Basic and diluted" )
Result:
output %>%
mutate(X1 = stringr::str_trunc(X1, 35))
#> # A tibble: 16 × 5
#> X1 X2 X3 X4 X5
#> <chr> <dbl> <chr> <chr> <chr>
#> 1 Care fees NA 210,187 194,603 398,206
#> 2 Management fees NA 59,746 50,959 105,552
#> 3 Interest received NA 364 42 41
#> 4 Other income NA 3,942 2,260 4,998
#> 5 Total revenue NA 274,239 247,864 508,797
#> 6 Fair-value movement of investmen... 3 261,346 285,143 745,885
#> 7 Total income NA 535,585 533,007 1,254,682
#> 8 Operating expenses NA (265,148) (225,380) (466,238)
#> 9 Depreciation and amortisation ex... NA (22,996) (17,854) (35,698)
#> 10 Finance costs NA (19,355) (15,250) (30,664)
#> 11 Impairment loss 2 (10,784) - -
#> 12 Total expenses NA (318,283) (258,484) (532,600)
#> 13 Profit before income tax NA 217,302 274,523 722,082
#> 14 Income tax (expense) / credit NA (23,316) 6,944 (29,209)
#> 15 Profit for the period NA 193,986 281,467 692,873
#> 16 Earnings per share Basic and dil... NA 38.8 56.3 138.6
Created on 2022-11-19 with reprex v2.0.2
I want to use the string str_pad function to make a column in my desired format, which includes zero padding the numbers in the "Code" column to 3 digits.
I've run this code:
Animals %>%
gather(most_common, cnt, M:OG) %>%
group_by(name) %>%
slice(which.max(cnt)) %>%
arrange(code)
Which resulted in the following tibble:
Code Name most_common
32 Monkey Africa
33 Wolf Europe
34 Tiger Asia
35 Godzilla Asia
#With 1 234 more rows
I'm happy with my code above. However, because I'm going to merge this df later on, I need the "Code" column to be three digits with zero padding (i.e. in the format "nnn" / "032"), as this:
Code Name most_common
032 Monkey Africa
033 Wolf Europe
034 Tiger Asia
035 Godzilla Asia
#With 1 234 more rows
I've run string str_pad($code, $3, $0), but it doesn't work. I guess there's something wrong there. Should I run this code wherever I want in my chunk or by using %>%?
A possible solution:
library(tidyverse)
df <- read.table(text = "Code Name most_common
32 Monkey Africa
33 Wolf Europe
34 Tiger Asia
35 Godzilla Asia", header = T)
df %>%
mutate(Code = str_pad(Code, width = 3, pad = "0"))
#> Code Name most_common
#> 1 032 Monkey Africa
#> 2 033 Wolf Europe
#> 3 034 Tiger Asia
#> 4 035 Godzilla Asia
In base R, we can use sprintf
df1$Code <- sprintf("%03d", df1$Code)
Another option could be using formatC with "d" for integer and a flag "0" the prepending zero like this:
df$Code <- formatC(df$Code, width = 3, format = "d", flag = "0")
df
#> Code Name most_common
#> 1 032 Monkey Africa
#> 2 033 Wolf Europe
#> 3 034 Tiger Asia
#> 4 035 Godzilla Asia
Created on 2022-07-23 by the reprex package (v2.0.1)
I've looked around for a while trying to figure this out, but I just can't seem to describe my problem concisely enough to google my way out of it. I am trying to work with Michigan COVID stats where the data has Detroit listed separately from Wayne County. I need to add Detroit's numbers to Wayne County's numbers, then remove the Detroit rows from the data frame.
I have included a screen grab too. For the purposes of this problem, can someone explain how I can get Detroit City added to Dickinson, and then make the Detroit City rows disappear? Thanks.
library(tidyverse)
library(openxlsx)
cases_deaths <- read.xlsx("https://www.michigan.gov/coronavirus/-/media/Project/Websites/coronavirus/Cases-and-Deaths/4-20-2022/Cases-and-Deaths-by-County-2022-04-20.xlsx?rev=f9f34cd7a4614efea0b7c9c00a00edfd&hash=AA277EC28A17C654C0EE768CAB41F6B5.xlsx")[,-5]
# Remove rows that don't describe counties
cases_deaths <- cases_deaths[-c(51,52,101,102,147,148,167,168),]
Code chunk output picture
You could do:
cases_deaths %>%
filter(COUNTY %in% c("Wayne", "Detroit City")) %>%
mutate(COUNTY = "Wayne") %>%
group_by(COUNTY, CASE_STATUS) %>%
summarize_all(sum) %>%
bind_rows(cases_deaths %>%
filter(!COUNTY %in% c("Wayne", "Detroit City")))
#> # A tibble: 166 x 4
#> # Groups: COUNTY [83]
#> COUNTY CASE_STATUS Cases Deaths
#> <chr> <chr> <dbl> <dbl>
#> 1 Wayne Confirmed 377396 7346
#> 2 Wayne Probable 25970 576
#> 3 Alcona Confirmed 1336 64
#> 4 Alcona Probable 395 7
#> 5 Alger Confirmed 1058 8
#> 6 Alger Probable 658 5
#> 7 Allegan Confirmed 24109 294
#> 8 Allegan Probable 3024 52
#> 9 Alpena Confirmed 4427 126
#> 10 Alpena Probable 1272 12
#> # ... with 156 more rows
Created on 2022-04-23 by the reprex package (v2.0.1)
I'm using the Google Trends R package to perform several queries of keywords like so:
trends1 <- gtrends(keyword="compare", gprop=channel,geo="AU", time=time, category=249)
trends2 <- gtrends(keyword="switch", gprop=channel,geo="AU", time=time, category=249)
trends3 <- gtrends(keyword="change", gprop=channel,geo="AU", time=time, category=249)
I'm only interested in the interest over time results, so I single them out:
time_trend1 <- trends1$interest_over_time
time_trend2 <- trends2$interest_over_time
time_trend3 <- trends3$interest_over_time
But I have 60 of these (and many more to add). I want to write a repeat loop (I think):
#select only interest over time
x <- 0
repeat {
time_trend(x+1) <- trends(x+1)$interest_over_time
if (x == 61){break}
}
but I get the error: Error in trends(x + 1) : could not find function "trends"
what am I missing?
You could use lapply to iterate over a list of keywords and extract the requested element like this:
library(gtrendsR)
time <- "today+5-y"
channel <- "web"
keywords <- list("compare", "switch", "change")
trends <- setNames(lapply(keywords, function(x) gtrends(keyword=x,
gprop=channel, geo="AU", time=time, category=249)), keywords)
lapply(trends, `[[`, "interest_over_time")
#> $compare
#> date hits geo time keyword gprop category
#> 1 2015-04-26 25 AU today+5-y compare web 249
#> 2 2015-05-03 26 AU today+5-y compare web 249
#> 3 2015-05-10 41 AU today+5-y compare web 249
#> 4 2015-05-17 29 AU today+5-y compare web 249
#> 5 2015-05-24 32 AU today+5-y compare web 249
# ...
#> 260 2020-04-12 9 AU today+5-y compare web 249
#>
#> $switch
#> date hits geo time keyword gprop category
#> 1 2015-04-26 0 AU today+5-y switch web 249
#> 2 2015-05-03 0 AU today+5-y switch web 249
#> 3 2015-05-10 0 AU today+5-y switch web 249
#> 4 2015-05-17 0 AU today+5-y switch web 249
#> 5 2015-05-24 0 AU today+5-y switch web 249
# ...
#> 260 2020-04-12 0 AU today+5-y switch web 249
#>
#> $change
#> date hits geo time keyword gprop category
#> 1 2015-04-26 45 AU today+5-y change web 249
#> 2 2015-05-03 68 AU today+5-y change web 249
#> 3 2015-05-10 23 AU today+5-y change web 249
#> 4 2015-05-17 52 AU today+5-y change web 249
#> 5 2015-05-24 76 AU today+5-y change web 249
# ...
#> 260 2020-04-12 38 AU today+5-y change web 249
Created on 2020-04-20 by the reprex package (v0.3.0)
Edit:
It may be easiest to further manipulate the data once the individual list elements get combined into a data.table, tibble or data.frame. Shown here is an example of how to remove unwanted columns. To subset by keyword, one could do, e.g. res[keyword=="compare"]
library(gtrendsR)
library(data.table)
time <- "today+5-y"
channel <- "web"
keywords <- list("compare", "switch", "change")
trends <- setNames(lapply(keywords, function(x) gtrends(keyword=x,
gprop=channel, geo="AU", time=time, category=249)), keywords)
res <- rbindlist(lapply(trends, `[[`, "interest_over_time"))
res[,-c("geo","category","time")]
#> date hits keyword gprop
#> 1: 2015-04-26 25 compare web
#> 2: 2015-05-03 26 compare web
#> 3: 2015-05-10 41 compare web
#> 4: 2015-05-17 29 compare web
#> 5: 2015-05-24 32 compare web
#> ---
#> 776: 2020-03-15 51 change web
#> 777: 2020-03-22 27 change web
#> 778: 2020-03-29 20 change web
#> 779: 2020-04-05 0 change web
#> 780: 2020-04-12 35 change web
Created on 2020-04-21 by the reprex package (v0.3.0)
You can get the data in a list using ls + mget, use lapply to iterate over each list and get "interest_over_time" element of each list.
total_list <- lapply(mget(ls(pattern = 'trends\\d+')), `[[`, "interest_over_time")
total_list would give you list of dataframes. It is better to keep data in a list since it easier to manage and does not clutter the environment with lot of objects. However, if you want data separately for each we can use list2env.
list2env(total_list, .GlobalEnv)
To drop certain columns, we can do :
total_list <- lapply(mget(ls(pattern = 'trends\\d+')), function(x) {
data <- x$interest_over_time
data[setdiff(names(data), c("geo","category","time"))]
})
Is it possible in some way to use a fit object, specifically the regression object I get form a plm() model, to flag observations, in the data used for the regression, if they were in fact used in the regression. I realize this could be done my looking for complete observations in my original data, but I am curious if there's a way to use the fit/reg object to flag the data.
Let me illustrate my issue with a minimal working example,
First some packages needed,
# install.packages(c("stargazer", "plm", "tidyverse"), dependencies = TRUE)
library(plm); library(stargazer); library(tidyverse)
Second some data, this example is drawing heavily on Baltagi (2013), table 3.1, found in ?plm,
data("Grunfeld", package = "plm")
dta <- Grunfeld
now I create some semi-random missing values in my data object, dta
dta[c(3:13),3] <- NA; dta[c(22:28),4] <- NA; dta[c(30:33),5] <- NA
final step in the data preparation is to create a data frame with an index attribute that describes its individual and time dimensions, using tidyverse,
dta.p <- dta %>% group_by(firm, year)
Now to the regression
plm.reg <- plm(inv ~ value + capital, data = dta.p, model = "pooling")
the results, using stargazer,
stargazer(plm.reg, type="text") # stargazer(dta, type="text")
#> ============================================
#> Dependent variable:
#> ---------------------------
#> inv
#> ----------------------------------------
#> value 0.114***
#> (0.008)
#>
#> capital 0.237***
#> (0.028)
#>
#> Constant -47.962***
#> (9.252)
#>
#> ----------------------------------------
#> Observations 178
#> R2 0.799
#> Adjusted R2 0.797
#> F Statistic 348.176*** (df = 2; 175)
#> ===========================================
#> Note: *p<0.1; **p<0.05; ***p<0.01
Say I know my data has 200 observations, and I want to find the 178 that was used in the regression.
I am speculating if there's some vector in the plm.reg I can (easily) use to crate a flag i my original data, dta, if this observation was used/not used, i.e. the semi-random missing values I created above. Maybe some broom like tool.
I imagine something like,
dta <- dta %>% valid_reg_obs(plm.reg)
The desired outcome would look something like this, the new element is the vector plm.reg at the end, i.e.,
dta %>% as_tibble()
#> # A tibble: 200 x 6
#> firm year inv value capital plm.reg
#> * <int> <int> <dbl> <dbl> <dbl> <lgl>
#> 1 1 1935 318 3078 2.80 T
#> 2 1 1936 392 4662 52.6 T
#> 3 1 1937 NA 5387 157 F
#> 4 1 1938 NA 2792 209 F
#> 5 1 1939 NA 4313 203 F
#> 6 1 1940 NA 4644 207 F
#> 7 1 1941 NA 4551 255 F
#> 8 1 1942 NA 3244 304 F
#> 9 1 1943 NA 4054 264 F
#> 10 1 1944 NA 4379 202 F
#> # ... with 190 more rows
Update, I tried to use broom's augment(), but unforunatly it gave me the error message I had hoped would create some flag,
# install.packages(c("broom"), dependencies = TRUE)
library(broom)
augment(plm.reg, dta)
#> Error in data.frame(..., check.names = FALSE) :
#> arguments imply differing number of rows: 200, 178
The vector is plm.reg$residuals. Not sure of a nice broom solution, but this seems to work:
library(tidyverse)
dta.p %>%
as.data.frame %>%
rowid_to_column %>%
mutate(plm.reg = rowid %in% names(plm.reg$residuals))
for people who use the class pdata.frame() to create an index attribute that describes its individual and time dimensions, you can us the following code, this is from another Baltagi in the ?plm,
# == Baltagi (2013), pp. 204-205
data("Produc", package = "plm")
pProduc <- pdata.frame(Produc, index = c("state", "year", "region"))
form <- log(gsp) ~ log(pc) + log(emp) + log(hwy) + log(water) + log(util) + unemp
Baltagi_reg_204_5 <- plm(form, data = pProduc, model = "random", effect = "nested")
pProduc %>% mutate(reg.re = rownames(pProduc) %in% names(Baltagi_reg_204_5$residuals)) %>%
as_tibble() %>% select(state, year, region, reg.re)
#> # A tibble: 816 x 4
#> state year region reg.re
#> <fct> <fct> <fct> <lgl>
#> 1 CONNECTICUT 1970 1 T
#> 2 CONNECTICUT 1971 1 T
#> 3 CONNECTICUT 1972 1 T
#> 4 CONNECTICUT 1973 1 T
#> 5 CONNECTICUT 1974 1 T
#> 6 CONNECTICUT 1975 1 T
#> 7 CONNECTICUT 1976 1 T
#> 8 CONNECTICUT 1977 1 T
#> 9 CONNECTICUT 1978 1 T
#> 10 CONNECTICUT 1979 1 T
#> # ... with 806 more rows
finally, if you are running the first Baltagi without index attributes, i.e. unmodified example from the help file, the code should be,
Grunfeld %>% rowid_to_column %>%
mutate(plm.reg = rowid %in% names(p$residuals)) %>% as_tibble()