Estimating Revenue for TH Hotel Brand Abuja using the Bilinear Auto Regression Moving Average (BARMA) Time Series Model

The increase in revenue within a destination reflects the quality of service delivery, especiallyin the hotel brand sector. This research explores the application of bilinear time series modelsto estimate the monthly revenue data of TH hotel brand Abuja. The study utilizes quantitativesecondary data, analysing revenue figures from 2017 to 2022. The aim was to identify andestimate a suitable bilinear time series model for the revenue data of the TH hotel brand up to2030. Both descriptive and inferential statistics were employed to test the hypothesis. TheAkaike Information Criterion (AIC) was used to determine the best-fitting model, revealing thatthe bilinear time series model (6, 0, 6, 0) is most suitable for modelling the revenue series.Additionally, the Shapiro-Wilk test for normality was conducted, showing that the nullhypothesis (H0) at an alpha level of 0.05 is rejected when the p-value is less than 0.05 (p <0.05). This indicates that the data tested do not come from a normally distributed population,highlighting the necessity of this research. The results demonstrate that Bilinear Fit Six(6,0,6,0) with standard error of 1.443, BIC of 1125.547, AIC of 1125.542, F- statistics of 6.769and P-value of 8.746*10-11 provides a superior model fit and forecasting accuracy comparedto other models. In contrast, the forecasting results suggest that the bilinear model predicts anincreasing trend in revenue up to 2030. This indicates that the bilinear time series model willlikely provide accurate forecasts, supporting rejecting the null hypothesis in favour of thealternative hypothesis