# Time series forecast for small data set

I am new in data science so please accept my apology in advance if my question sounds stupid. I want to do a time series forecast of outage mins in the current regulatory year. The regulatory year starts from 1 April and ends on 30 March of next year. I have data of around six months i.e. from April to September. Outage does not occur every day. So I have only 144 data points (or days out of 171 days) where the outage occurred. I have plotted the data in the following graph. The graph shows the cumulative sum of outage mins.

Now I am trying to predict the value from October to March. I wanted to forecast the value that what would be the cumulative outage mins by the end of March next year. I tried to use Exponential smoothing but it did not work, it may be because I don't have a lot of observation. Then I was reading about ARIMA but not sure whether its the right algorithm to use or not as I don't think that there would be any seasonality in this scenario and also I don't have long data points. Could anyone help with which algorithm should I use to forecast the value? I am using Python as a programming language. Any help would be really appreciated.

ARIMA could work, I think it's the right approach. It's simple enough to be used on a small dataset, but sufficiently flexible at the same time. If you are using Python, library statsmodels allows you to implement ARIMA regressions. You have to grid search and find the right parameters to find the best fit, and run the prediction.
• There is a method, something like model.conf_int(), that will get you confidence intervals. – Leevo Sep 23 '20 at 7:26