forecasting multiple time series in R using auto.arima

Question

my dataset has the following 3 columns:

date client_id sales
01/01/2012 client 1 $1000
02/01/2012 client 1 $900
...
...
12/01/2014 client 1 $1000
01/01/2012 client 2 $300
02/01/2012 client 2 $450
...
..
12/01/2014 client 2 $375

and so on for 98 other clients (24 monthly datapoints for each client)

I have multiple clients (around 100)...data is in time series format for each client (24 monthly datapoints)

how do I automatically forecast sales for all the 100 clients using auto.arima in R? is there a by statement option? or do i have to use loops?

Thanks

Answer 1

You can always use lapply():

lapply(tsMat, function(x) forecast(auto.arima(x)))

A little example follows:

library(forecast)
#generate some time-series:
sales <- replicate(100, 
    arima.sim(n = 24, list(ar = c(0.8), ma = c(-0.2)), sd = sqrt(0.1)) 
)
dates <- seq(as.Date("2012/1/1"), by = "month", length.out=24)
df <- data.frame(date=rep(dates,100), client_id=rep(1:100,each=24), sales=c(sales))
#reshape and convert it to a proper time-series format like ts:
tsMat <- ts(reshape2::dcast(df, date~client_id), start=2012, freq=12)
#forecast by auto.arima:
output <- lapply(tsMat, function(x) forecast(auto.arima(x)))

Answer 2

You can also specify the number to forecast in the future by using 'h=#ofPeriods' in the forecast call

Forecast.allStates <- as.data.frame(lapply(ts.allStates, function(x) forecast(auto.arima(x),h=67)))