Forecasting is an essential subject in economics and business. Autoregressive Integrated Moving Average (ARIMA) has been extensively used despite its weaknesses, from requiring a minimum number of data points to the assumed linearity of data, which are not always feasible in business data. Therefore, ARIMA is unable to predict well with volatile data. With recent advancements, the Long Short-Term Memory (LSTM) shows potential to address such weaknesses. This research is aimed to identify the effect of the number of short and long data points of the time series data, as well as a model that is more suitable for handling volatile data due to missing values. Performance metrics used are the model accuracy measured with RMSE and the model run-time performance measured with the Python Timeit library. This research concluded that LSTM (RMSE 0.037618) is more accurate than ARIMA (RMSE 0.062667) in a dataset of 60 data points and it is reverted in a longer dataset of 228 data points (RMSE of ARIMA 0.006949 to LSTM 0.036025). While in the case of missing values, LSTM outperforms ARIMA, although both models have decreased accuracy if the number of missing values is increased. In terms of run-time performance, ARIMA is significantly faster than LSTM.