May some use the attached R script on attached data and let me help interpretation of the output please . In data and R code , X = number of outbreaks per month of X disease in a county (data from July 2010 to December 2014)
Regards
Dear Tariq,
can you upload more data? i am afraid that 54 observations are not enough!
i mean you can not trust the stationarity tests that you used. in the DECOMPOSE variable, we need to check the 'random' factor whether it's stationary. the ACF plot suggest that X is barely correlated to its past value. the results of Auto.arima show a MAE of 1.4556. this is not really acceptable accuracy because the mean 50% of X range between 0 and 2.00 and the variance of X is 3.569.
to summarize: if you are going to make forecasting based on this data only; i am not sure how far can you go with this!!. if you have more data or other variable please share it with us.
regards;
try this, it might be useful.
http://www.rdatamining.com/examples/time-series-analysis
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series