I have a data frame like this called "fulldata" with over 12000 observations:
Station_id Date Prec_daily
44 2019-04-30 0.5
172 2019-05-21 0.0
82 2019-04-30 2.3
44 2019-05-07 4.7
250 2019-05-21 0.0
45 2019-05-02 NA
890 2019-04-30 0.5
82 2019-05-14 5.2
250 2019-05-07 NA
(Station_id = integer, Date = POSIXct, Prec_daily = numeric)
Station_id's stand for different weather stations across a country. The variable precipitation_daily shows the daily amount of rainfall on the given date at the given station. The dates are days on which a voluntary nature observation was submitted to a platform by a participant of the project.
My goal is being able to make a prediction like "When I have precipitation of 1.5 in one day, there will be an estimated amount of XX observations that day (across the country)."
How can I do this in R? I think I need a new data frame or variable that calculates the average amount of total observations across the country for every amount of rainfall in a day.
I really struggle with that right now. So far my analysis contains a histogram of the df above and the mean of the Prec_daily variable. I also managed to make a new data frame in which the total frequency for every amount of rainfall is counted. For that I used the following code.
PrecClean <- fulldata$Prec_daily[!is.na(fulldata$Prec_daily)]
hist(PrecClean, xlim = c(0,10), ylim = c(0, 11000))
box()
mean(PrecClean)
Precfreq <- count(fulldata, vars = "Prec_daily")
But that doesn't help me determining the estimated DAILY amount of observations for each precipitation level...
Thanks a lot in advance for any advice!
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…