CODEWITHSUNDEEP
rarima
Ojaswita
Ojaswita

Reputation: 93

Calculating the RMSE and ACF plot of residuals of ARIMA model in R

I have a dataset labeled covid19india. I used a dataset to forecast the number of cases using the following code and got the graph as desired. I am ok up to this point, but need help in calculating the RMSE and plotting the ACF plot of residuals to theoretically show that the model is feasible. 

set1 <- covid19india[1:36,]
df <- set1[3]
tddf1 <- ts(df$`Cumulative cases`)
fit1 <- auto.arima(df$`Cumulative cases`, seasonal = FALSE)
forecast3 <- forecast(fit1, h=9)
plot(forecast3)
par(new=TRUE)
plot(days, df1)

I need help in calculating the RMSE and residual ACF plot.

Upvotes: 1

Views: 2281

Answers (1)

Earl Mascetti
Earl Mascetti

Reputation: 1336

You should use the function checkresiduals presents in the forecast package.

Below a simple example.

    >library(forecast)
    >fit_1<-auto.arima(your_data_set)
    >forecast(fit_1, h = 10) # h is the period that you want to forecast. 
    >checkresiduals(fit_1)

enter image description here

To check instead the RMSE you could use the function accuracy

> accuracy(fit_1)
                       ME       RMSE        MAE  MPE MAPE      MASE        ACF1
Training set 2.236275e-05 0.02440796 0.01829821 -Inf  Inf 0.8858579 -0.01149095

If you'd like to deep you should check this

This link is the free book that the Prof. Rob J Hyndman and Prof George Athanasopoulos wrote (are the authors of the forecast package).

Upvotes: 1

Related Questions