This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
I've to forecast the daily data based on ARIMA and ETS data(use the best based on TS Compare). I've 20 column fields. I've to forecast the 'id_subzone' field for next two months (daily all day) based on previous data. But, when I'm forecasting for 'id_subzone' field, I'm getting only few records forecasted. I'm attaching my alteryx workflow below.
How to customize ARIMA model for my' tiz_data.yxdb' for better accuracy of prediction.
I was able to successfully forecast 120 periods with the ETS tool in my Designer (2018.4). Can you double check you put 120 periods? Are there any notes in the log?
On another note, your forecast appears to be a straight line. You are likely using an Additive Error Type, with no Trend, Trend dampening, or Seasonality. This means that your Confidence Interval just gets larger with each period, but the forecasts are all the same. You may want to change your settings or possibly look at ARIMA.
See the screenshot on my side. Note you will likely need a lot of data to predict 120 periods into the future.