A Comparative study Statistical and Machine Learning Models for Monthly Airline Passenger Forecasting
Computer and Decision Making: An International Journal · 05.11.2026 · 3 (6S) · 2664-2671
DOI: 10.1922/comdem.v3i6s.19004 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 05.11.2026. Verfasst von Kalyan Kumar Palaparthi und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Forecasting methods should be chosen based on out-of-sample evaluation, not by considering the complexity of the model. This study benchmarks a seasonal naïve model, linear regression with a seasonal component, random forest and gradient boosting with the public Air Passengers data set. There are 144 observations in the data set, one for each month from January 1949 to December 1960. A lag 1, lag 12, a time index and calendar indicators are used to fit supervised models for 120 target months January 1950 to December 1959. The fitted parameters are kept constant and twelve sequential one month-ahead forecasts are checked during 1960. The seasonal naïve benchmark model has the highest mean absolute error of 47.83 thousand passengers, followed by gradient boosting at 22.64, random forest at 25.89 and seasonal regression. The mean absolute error was reduced by 65.53% about the benchmark. It has a root squared error of 20.64 and a mean absolute scaled error of 0.542 for the seasonal periods. The summer peak is difficult for tree ensembles to predict, as shown in monthly predictions. The results here confirm that this series and protocol have an interpretable regression model, but do not offer any grounds for a general ranking of the different statistical and machine learning methods. The study provides an audit trail computational comparison with explicit feature timing, model parameters and constraints. In this paper, the authors employ time series forecasting methods to forecast the number of passengers on the AirPassengers dataset and evaluate the results.
Zitieren
APA
Palaparthi, K. K., Chary, B. V., Bhalerao, P., Reddy, Y., Sivaiah, S., & Efana, M. (2026). A Comparative study Statistical and Machine Learning Models for Monthly Airline Passenger Forecasting. Computer and Decision Making: An International Journal, 3(6S), 2664-2671. https://doi.org/10.1922/comdem.v3i6s.19004
Vancouver
Palaparthi KK, Chary BV, Bhalerao P, Reddy Y, Sivaiah S, Efana M. A Comparative study Statistical and Machine Learning Models for Monthly Airline Passenger Forecasting. Computer and Decision Making: An International Journal. 2026;3(6S):2664-2671. doi:10.1922/comdem.v3i6s.19004
BibTeX
@article{Palaparthi2026,
author = {Palaparthi, Kalyan Kumar and Chary, B Vasudeva and Bhalerao, Priyanka and Reddy, YMadhavi and Sivaiah, S and Efana, MD},
title = {A Comparative study Statistical and Machine Learning Models for Monthly Airline Passenger Forecasting},
journal = {Computer and Decision Making: An International Journal},
year = {2026},
volume = {3},
number = {6S},
pages = {2664--2671},
doi = {10.1922/comdem.v3i6s.19004}
}
EUVIMED