Please check the quality of forecast in "Describe Forecast" option, you will surely understand the trend over there if the quality is no good and in case if quality of Forecast is Good then I guess the behavior is as expected. However if you want to cross check my suggestion would be to change the database values and observe the Trend.
Describe forecast.png 38.4 KB
Part of the issue may be that there is not enough data. The following information may give some more insight:
Must have good data and the correct model for your data.
- If the quality of your forecast is poor, the process generating your data might not match the four models available in Tableau or the random variance in your data might be too high. In this case, you cannot trust the Precision of your forecast to be correct.
- For example, the seasonal swings in your sales may be impacted by the amount of sales. But, the seasonal component of the current Tableau model assumes these factors are independent. So, in this case, you may get a better forecast if you force Tableau not to try to model seasonality.