问题
I am trying to do forecasting of stock prices in R using ARIMA. I am using the auto.arima function to fit my model. Every time I'm trying to do that I get the same value for the forecasted values. I tried using different stocks but the same thing happens in every case. Here I tried forecasting apple prices:
arimapple <- ts(appletrain, start = timedata[1])
fitappletrain <- auto.arima(arimapple)
fitappletrain
forecastapple <- forecast(fitappletrain, h=57)
forecastapple
and the output that I get is as follows:
Point Forecast Lo 80 Hi 80 Lo 95 Hi 95
17763 180.94 176.7350 185.1450 174.5090 187.3710
17764 180.94 174.9932 186.8868 171.8451 190.0349
17765 180.94 173.6567 188.2233 169.8011 192.0789
17766 180.94 172.5299 189.3501 168.0779 193.8021
17767 180.94 171.5373 190.3427 166.5598 195.3202
17768 180.94 170.6398 191.2402 165.1872 196.6928
17769 180.94 169.8145 192.0655 163.9251 197.9549
17770 180.94 169.0464 192.8336 162.7503 199.1297
17771 180.94 168.3249 193.5551 161.6469 200.2331
and so on, so every forecast I receive is 180.94. How can I solve this?
回答1:
I think your data does not contain any strong seasonality and trend and i think your historical data is also less, the auto.arima() is not able to find forecast accurate because of that.
Therefore it is simply taking average of your historical data and returning as forecast. therefore the values are same (you will get straight line when you plot it.)
来源：https://stackoverflow.com/questions/54842872/same-values-for-arima-forecasts-in-r