Forecasting The Outbursts Of The Photometry Light Curve Of Star V363 Lyr

In this paper we investigate the astronomical time series (TS) which is the photometric observations of the variable object V363 Lyr. We perform the spectral analysis of the time series and compare two approaches to forecast the outbursts of this time series. Since the data contain missing values we do the missing values imputation as well. The outbursts occur regularly, but our analysis shows that they are not strictly periodic. Hence, to improve the forecast of outburst position we compare several machine learning techniques for the two main forecasting approaches. Our results show that each approach can be beneficial depending on the starting point of forecast.