I have monthly price data for tomatoes for the last 9 yrs for a particular town and I'm looking to predict the prices of tomatoes 6 months into the future.

I had considered using Linear Regression in Tensorflow (something I learnt about a week ago), because there is a direct relationship between price and rainfall for the location in question. That is, high rainfall means a supply glut and a price drop, low rainfall means scarcity and price hikes.

However I've found that I can't get accurate weather forecasts for 6 months into the future and have to use some other way other than Linear Regression to predict future tomato prices.

What is the best way to do this?


closed as too broad by Siong Thye Goh, Toros91, Mark.F, oW_, Sean Owen Feb 5 at 22:50

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.


You can find the seasonality in your data (if there exist) and by removing the seasonality, find the trend of tomato price and predict the future price using the seasonality data and trend data. If you are using python, you can find the seasonal library useful to apply this method to your data.

The advantage of this method to your method is you don't need to think over finding dependent features and any feature engineering.

Moreover, as you have mentioned in tags of your question, you can using RNN to predict the future value by as your data is time series and sequential intrinsically. You can find this article useful.

  • $\begingroup$ Thanks! I'm taking a look at the resources you provided. Some light research pointed me at RNNs though I don't know anything about them. Thanks for the links, will take a look at them. $\endgroup$ – Jonathan Feb 3 at 18:43
  • $\begingroup$ Quick question, why would I want to remove seasonality from the data? $\endgroup$ – Jonathan Feb 3 at 18:44
  • $\begingroup$ @Jonathan remove the seasonality to find the trend. However, you need both to predict the future price. $\endgroup$ – OmG Feb 4 at 9:40

Not the answer you're looking for? Browse other questions tagged or ask your own question.