I am having a daily dataset of 4 parameters which I have converted into monthly data using following code
library(zoo)
library(hydroTSM)
library(lubridate)
library(tidyverse)
set.seed(123)
df <- data.frame("date"= seq(from = as.Date("1983-1-1"), to = as.Date("2018-12-31"), by = "day"),
"Parameter1" = runif(length(seq.Date(as.Date("1983-1-1"), as.Date("2018-12-31"), "days")), 15, 35),
"Parameter2" = runif(length(seq.Date(as.Date("1983-1-1"), as.Date("2018-12-31"), "days")), 11, 29),
"Parameter3" = runif(length(seq.Date(as.Date("1983-1-1"), as.Date("2018-12-31"), "days")), 50, 90),
"Parameter4" = runif(length(seq.Date(as.Date("1983-1-1"), as.Date("2018-12-31"), "days")), 0, 27))
Monthly_data <- daily2monthly(df, FUN=mean, na.rm=TRUE)
After that, I have reshaped it to represent each column as month using following code
#Function to convert month abbreviation to a numeric month
mo2Num <- function(x) match(tolower(x), tolower(month.abb))
Monthly_data %>%
dplyr::as_tibble(rownames = "date") %>%
separate("date", c("Month", "Year"), sep = "-", convert = T) %>%
mutate(Month = mo2Num(Month))%>%
tidyr::pivot_longer(cols = -c(Month, Year)) %>%
pivot_wider(names_from = Month, values_from = value, names_prefix = "Mon",
names_sep = "_") %>%
arrange(name)
Now, I want to create parameter combinations like Parameter1 * Parameter2, Parameter1 * Parameter3, Parameter1 * Parameter4, Parameter2 * Parameter3, Parameter2 * Parameter4, Parameter3 * Parameter4 which will be added to the pivoted monthly data as rbind. The new dataframe Parameter1 * Parameter2 means to multiply their monthly values and then rbind to the above result. Likewise for all other above said combinations. How can I achieve this?
You can use this base R approach using combn assuming data is present for all the years for all parameters where df1 is the dataframe from the above output ending with arrange(name).
data <- combn(unique(df1$name), 2, function(x) {
t1 <- subset(df1, name == x[1])
t2 <- subset(df1, name == x[2])
t3 <- t1[-(1:2)] * t2[-(1:2)]
t3$name <- paste0(x, collapse = "_")
cbind(t3, t1[1])
}, simplify = FALSE)
You can then rbind it to original data.
new_data <- rbind(df1, do.call(rbind, data))
Related
In R I've got a dataset like this one:
df <- data.frame(
ID = c(1:30),
x1 = seq(0, 1, length.out = 30),
x2 = seq(100, 3000, length.out = 30),
category = gl(3, 10, labels = c("NEGATIVE", "NEUTRAL", "POSITIVE"))
)
Now I want to add a new column with randomized boolean values, but inside each category the proportion of TRUE and FALSE values should be the same (i.e. the randomizing process should generate the same count of true and false values, in the above data frame 5 TRUEs and 5 FALSEs in each of the 3 categories). How to do this?
You can sample a vector of "TRUE" and "FALSE" values without replacement so you have a randomized and balanced column in your data-frame.
sample(rep(c("TRUE","FALSE"),each=5),10,replace=FALSE)
Based on Yacine Hajji answer:
addRandomBool <- function(df, p){
n <- ceiling(nrow(df) * p)
df$bool <- sample(rep(c("TRUE","FALSE"), times = c(n, nrow(df) - n)))
df
}
Reduce(rbind, lapply(split(df, df$category), addRandomBool, p = 0.5))
where parametar p determines the proportion of TRUE.
This will sample within each group from a vector of 5 TRUE and 5 FALSE without replacement. It will assume that there are always 10 records per group.
library(dplyr)
library(tidyr)
df <- data.frame(
ID = c(1:30),
x1 = seq(0, 1, length.out = 30),
x2 = seq(100, 3000, length.out = 30),
category = gl(3, 10, labels = c("NEGATIVE", "NEUTRAL", "POSITIVE"))
)
set.seed(pi)
df %>%
group_by(category) %>%
nest() %>%
mutate(data = lapply(data,
function(df){ # Function to saple and assign the new_col
df$new_col <- sample(rep(c(FALSE, TRUE),
each = 5),
size = 10,
replace = FALSE)
df
})) %>%
unnest(cols = "data")
This next example is a little more generalized, but still assumes (approximately) even distribution of TRUE and FALSE within a group. But it can accomodate variable group sizes, and even groups with odd numbers of records (but will favor FALSE for odd numbers of records)
library(dplyr)
library(tidyr)
df <- data.frame(
ID = c(1:30),
x1 = seq(0, 1, length.out = 30),
x2 = seq(100, 3000, length.out = 30),
category = gl(3, 10, labels = c("NEGATIVE", "NEUTRAL", "POSITIVE"))
)
set.seed(pi)
df %>%
group_by(category) %>%
nest() %>%
mutate(data = lapply(data,
function(df){
df$new_col <- sample(rep(c(FALSE, TRUE),
length.out = nrow(df)),
size = nrow(df),
replace = FALSE)
df
})) %>%
unnest(cols = "data")
Maintaining Column Order
A couple of options to maintain the column order:
First, you can save the column order before you do your group_by - nest, and then use select to set the order when you're done.
set.seed(pi)
orig_col <- names(df) # original column order
df %>%
group_by(category) %>%
nest() %>%
mutate(data = lapply(data,
function(df){
df$new_col <- sample(rep(c(FALSE, TRUE),
length.out = nrow(df)),
size = nrow(df),
replace = FALSE)
df
})) %>%
unnest(cols = "data") %>%
select_at(c(orig_col, "new_col")) # Restore the column order
Or you can use a base R solution that doesn't change the column order in the first place
df <- split(df, df["category"])
df <- lapply(df,
function(df){
df$new_col <- sample(rep(c(FALSE, TRUE),
length.out = nrow(df)),
size = nrow(df),
replace = FALSE)
df
})
do.call("rbind", c(df, list(make.row.names = FALSE)))
There are likely a dozen other ways to do this, and probably more efficient ways that I'm not thinking of.
I have created a function which completes a calculation based on data from two data frames for an individual.
I want to complete that function for each individual and combine all the outputs in a data frame and export to .csv
Currently the output .csv only has data for person 34, none of the other.
I've noted that when I run this it creates an object ID, which is just the numeric 34. It seems to be only holding on to the last ID in data$ID.
How can I create an output with results for all persons?
library(dplyr)
library(lubridate)
library(date)
screen_function = function(x){
# Select each person and get necessary inputs
data = data%>%filter(ID == x)
demogs = demogs %>% filter (P_ID == x)
data$Age = demogs$Age
data$result = data$test * data$Age
data$Date = as.Date(data$Date,format='%d/%m/%Y') # ensures date column is in correct format
# only include tests from most recent 12-24 months and only proceed if test in last 12 months
Recent_12m = data %>% filter(between(Date, today() - years(1), today()))
Recent_24m = data %>% filter(between(Date, today() - years(2), today()))
if ((nrow(Recent_12m)) > 0) {
data = rbind(Recent_12m,Recent_24m)
Recent_12m$min_date = min(Recent_12m$Date)
Recent_12m$Date = as.Date(Recent_12m$Date,format='%d/%m/%Y')
Recent_12m$min_date = as.Date(Recent_12m$min_date,format='%d/%m/%Y')
Recent_24m$min_date = min(Recent_24m$Date)
Recent_24m$Date = as.Date(Recent_24m$Date,format='%d/%m/%Y')
Recent_24m$min_date = as.Date(Recent_24m$min_date,format='%d/%m/%Y')
Recent_12m$Period = interval(Recent_12m$min_date, Recent_12m$Date)
Recent_12m$Years = as.numeric(Recent_12m$Period, unit = "years")
Recent_24m$Period = interval(Recent_24m$min_date, Recent_24m$Date)
Recent_24m$Years = as.numeric(Recent_24m$Period, unit = "years")
# Latest result
Last = filter(Recent_12m, (Recent_12m$Date == max(Date)))
# linear regression model
Reg_12month <- lm(result ~ Years, data=Recent_12m)
Reg_24month <- lm(result ~ Years, data=Recent_24m)
info = c(x, round(Last$result, digits=1), round(Reg_12month$coefficients["Years"], digits = 1), round(Reg_24month$coefficients["Years"], digits = 1))
newdf = data.frame(matrix(0, ncol = 4))
colnames(newdf) = c("ID", "Latest result", "Trend 12month", "Trend 24 month")
newdf= rbind(newdf, info)
write.csv(newdf, "filepath.csv")
}
}
Date= sample(seq(as.Date('2019/11/01'), as.Date('2020/11/01'), by="day"), 12)
ID= c(12,12,12,450,450,450,1,1,1,34,34,34)
test= rnorm(12, mean=150, sd=60)
data= data.frame(ID, Date, test)
P_ID = c(1,12,34,450)
Age = c(50,45,60,72)
demogs = data.frame(P_ID, Age)
persons = unique(data$ID)
for(ID in persons){
screen_function(paste("", ID,"", sep=""))
}
Created on 2020-11-16 by the reprex package (v0.3.0)
So, I've got around this by using a pre-made .csv, instead of creating a new dataframe. The .csv just contains a single row of 4 columns, with random entries in each cell.
newdf= read.csv(file = "filepath.csv")
info = c(x, round(Last$result, digits=1), round(Reg_12month$coefficients["Years"], digits = 1),
round(Reg_24month$coefficients["Years"], digits = 1))
newdf= rbind(Summary, patient_info)
colnames(newdf) = c("ID", "Latest result", "Trend 12month", "Trend 24 month")
newdf= distinct(newdf, ID, .keep_all = TRUE)
write.csv(Summary, "filepath.csv", row.names = FALSE)}}
I have the following data table:
library(dplyr)
set.seed(123)
dt <- data.table(date = seq(as.Date('2020-01-01'), by = '1 day', length.out = 365),
Germany = rnorm(365, 2, 1), check.names = FALSE)
dt <- dt %>%
mutate(month = format(date, '%b'),
date = format(date, '%d')) %>%
tidyr::pivot_wider(names_from = date, values_from = Germany)
I would like to add two new columns (monthlyAverage, quarterlyAverage), one containing the monthly averages and the other column the quarterly averages.
For monthly average you can take rowwise mean, for quaterly average you can create groups of 3 rows and take mean of every 3 months.
library(dplyr)
dt %>%
mutate(monthlyaverage = rowMeans(.[-1], na.rm = TRUE)) %>%
group_by(grp = ceiling(row_number()/3)) %>%
mutate(quaterlyaverage = mean(monthlyaverage)) %>%
select(month, grp, monthlyaverage, quaterlyaverage, everything())
If you want to do this using data.table :
library(data.table)
setDT(dt)[, monthlyaverage := rowMeans(.SD, na.rm = TRUE), .SDcols = -1]
dt[, quaterlyaverage := mean(monthlyaverage), ceiling(seq_len(nrow(dt))/3)]
I want to replace Jan 01 to Jun 25 of all the years in FakeData with data from Ob2020 for the two variables (Level & Flow) of my data.frame. Here is what i have started and am looking for suggestions to achieving my goal.
library(tidyverse)
library(lubridate)
set.seed(1500)
FakeData <- data.frame(Date = seq(as.Date("2010-01-01"), to = as.Date("2018-12-31"), by = "days"),
Level = runif(3287, 0, 30), Flow = runif(3287, 1,10))
Ob2020 <- data.frame(Date = seq(as.Date("2020-01-01"), to = as.Date("2020-06-25"), by = "days"),
Level = runif(177, 0, 30), Flow = runif(177, 1,10))
Here's a way using dplyr and lubridate :
library(dplyr)
library(lubridate)
FakeData %>%
mutate(day = day(Date), month = month(Date)) %>%
left_join(Ob2020 %>%
mutate(day = day(Date), month = month(Date)),
by = c('day', 'month')) %>%
mutate(Level = coalesce(Level.y, Level.x),
Flow = coalesce(Flow.y, Flow.x)) %>%
select(Date = Date.x, Level, Flow)
If you dont mind a data.table solution, here is an update join:
library(data.table)
#extract year and month of the date
setDT(FakeData)[, c("day", "mth") := .(mday(Date), month(Date))]
setDT(Ob2020)[, c("day", "mth") := .(mday(Date), month(Date))]
#print to console to show old values
head(FakeData)
head(Ob2020)
cols <- c("Level", "Flow")
FakeData[Ob2020[mth<=6L & day<=25], on=.(day, mth),
(cols) := mget(paste0("i.", cols))]
#print to console to show new values
head(FakeData)
I am trying to compute the upper and lower quartile of the two variables in my data.frame across the time period of my interest. The code below gave me single digit for upper and lower value.
set.seed(50)
FakeData <- data.frame(seq(as.Date("2001-01-01"), to= as.Date("2003-12-31"), by="day"),
A = runif(1095, 0,10),
D = runif(1095,5,15))
colnames(FakeData) <- c("Date", "A","D")
statistics <- FakeData %>%
gather(-Date, key = "Variable", value = "Value") %>%
mutate(Year = year(Date), Month = month(Date)) %>%
filter(between(Month,3,5)) %>%
mutate(NewDate = ymd(paste("2020", Month,day(Date), sep = "-"))) %>%
group_by(Variable, NewDate) %>%
summarise(Upper = quantile(Value,0.75, na.rm = T),
Lower = quantile(Value, 0.25, na.rm = T))
I would want an output like below (the Final_output is what i am interested)
Output1 <- data.frame(seq(as.Date("2000-03-01"), to= as.Date("2000-05-31"), by="day"),
Upper = runif(92, 0,10), lower = runif(92,5,15), Variable = rep("A",92))
colnames(Output1)[1] <- "Date"
Output2 <- data.frame(seq(as.Date("2000-03-01"), to= as.Date("2000-05-31"), by="day"),
Upper = runif(92, 2,10), lower = runif(92,5,15), Variable = rep("D",92))
colnames(Output2)[1] <- "Date"
Final_Output<- bind_rows(Output1,Output2)
I can propose you a data.table solution. In fact there are several ways to do that.
The final steps (apply quartile by group on the Value variable) could be translated into (if you want, as in your example, two columns):
statistics[,.('p25' = quantile(get('Value'), probs = 0.25), 'p75' = quantile(get('Value'), probs = 0.75)),
by = c("Variable", "NewDate")]
If you prefer long-formatted output:
library(data.table)
setDT(statistics)
statistics[,.(lapply(get('Value'), quantile, probs = .25,.75)) ,
by = c("Variable", "NewDate")]
All steps together
It's probably better if you chose to use data.table to do all steps using data.table verbs. I will assume your data have the structure similar to the dataframe you generated and arranged, i.e.
statistics <- FakeData %>%
gather(-Date, key = "Variable", value = "Value")
In that case, mutate and filter steps would become
statistics[,`:=`(Year = year(Date), Month = month(Date))]
statistics <- statistics[Month %between% c(3,5)]
statistics[, NewDate = :ymd(paste("2020", Month,day(Date), sep = "-"))]
And choose the final step you prefer, e.g.
statistics[,.('p25' = quantile(get('Value'), probs = 0.25), 'p75' = quantile(get('Value'), probs = 0.75)),
by = c("Variable", "NewDate")]