Data Science Stack Exchange is a question and answer site for Data science professionals, Machine Learning specialists, and those interested in learning more about the field. It only takes a minute to sign up.
Sign up to join this community
Doing time series analysis, I have doubts on choosing the right model. I want to predict the next 30 mins window, from the input dataset which contains the no. of error count for that particular 1 min interval.
Should I use XGBoost or ARIMA regressions?
Most of article or tutorial I found online use ARIMA for time series, while XGBoost is used more in Kaggle competitions.
Usually, ARIMA regressions are used in classical statistical approaches, when the goalis not just prediction, but also understanding on how different explanatory variables relate with the dependent variable and with each other. ARIMA are thought specifically for time series data.
On the contrary, XGBoost models are used in pure Machine Learning approaches, where we exclusively care about quality of prediction. XGBoost regressors can be used for time series forecast (an example is this Kaggle kernel), even though they are not specifically meant for long term forecasts. But they can work.
In conclusion, I don't know your data, but if I had to bet I'd go for ARIMA. However both could work great, and I suggest you to try both and pick the best for your particular need. They are comparatively quick to implement, it shouldn't take too much time.
