Forecasting air temperatures using time series models and neural-based algorithms

Mahmoud M. Smadi, Farouq S. Mjalli

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Recently, neural network models have been popular and found useful for forecasting a wide variety of time series data in many disciplines. This is due to their favorable modeling properties of simplicity, fault and noise tolerance and their capability to adapt to process changes. Nevertheless, applications in climatology have been less widespread than other disciplines such as economics. In this study, feed-forward neural-network (FFNN) and autoregression (AR) time series models are used in forecasting the annual air temperature time series data in Jordan. The performance of the two predictors was compared using out-of-sample forecasts. The test period was shifted through the whole available time. As demonstrated by the forecasting experiments, the FFNN models gave better forecasts and were able to identify the dynamics of the temperature time series and gave more realistic forecasts. Both predictors showed a cooling trend in annual air temperatures for the coming 10 years.

Original languageEnglish
Pages (from-to)44-48
Number of pages5
JournalJournal of Mathematics and Statistics
Volume3
Issue number2
Publication statusPublished - 2007

Fingerprint

Time Series Models
Forecast
Forecasting
Feedforward Neural Networks
Time Series Data
Neural Network Model
Annual
Predictors
Autoregression
Tolerance
Cooling
Simplicity
Fault
Time series
Economics
Modeling
Experiment

Keywords

  • Air temperature
  • Autoregression
  • Forecasting
  • Neural network

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

Forecasting air temperatures using time series models and neural-based algorithms. / Smadi, Mahmoud M.; Mjalli, Farouq S.

In: Journal of Mathematics and Statistics, Vol. 3, No. 2, 2007, p. 44-48.

Research output: Contribution to journalArticle

@article{de62c2d86b1a493285c3943215684439,
title = "Forecasting air temperatures using time series models and neural-based algorithms",
abstract = "Recently, neural network models have been popular and found useful for forecasting a wide variety of time series data in many disciplines. This is due to their favorable modeling properties of simplicity, fault and noise tolerance and their capability to adapt to process changes. Nevertheless, applications in climatology have been less widespread than other disciplines such as economics. In this study, feed-forward neural-network (FFNN) and autoregression (AR) time series models are used in forecasting the annual air temperature time series data in Jordan. The performance of the two predictors was compared using out-of-sample forecasts. The test period was shifted through the whole available time. As demonstrated by the forecasting experiments, the FFNN models gave better forecasts and were able to identify the dynamics of the temperature time series and gave more realistic forecasts. Both predictors showed a cooling trend in annual air temperatures for the coming 10 years.",
keywords = "Air temperature, Autoregression, Forecasting, Neural network",
author = "Smadi, {Mahmoud M.} and Mjalli, {Farouq S.}",
year = "2007",
language = "English",
volume = "3",
pages = "44--48",
journal = "Journal of Mathematics and Statistics",
issn = "1549-3644",
publisher = "Science Publications",
number = "2",

}

TY - JOUR

T1 - Forecasting air temperatures using time series models and neural-based algorithms

AU - Smadi, Mahmoud M.

AU - Mjalli, Farouq S.

PY - 2007

Y1 - 2007

N2 - Recently, neural network models have been popular and found useful for forecasting a wide variety of time series data in many disciplines. This is due to their favorable modeling properties of simplicity, fault and noise tolerance and their capability to adapt to process changes. Nevertheless, applications in climatology have been less widespread than other disciplines such as economics. In this study, feed-forward neural-network (FFNN) and autoregression (AR) time series models are used in forecasting the annual air temperature time series data in Jordan. The performance of the two predictors was compared using out-of-sample forecasts. The test period was shifted through the whole available time. As demonstrated by the forecasting experiments, the FFNN models gave better forecasts and were able to identify the dynamics of the temperature time series and gave more realistic forecasts. Both predictors showed a cooling trend in annual air temperatures for the coming 10 years.

AB - Recently, neural network models have been popular and found useful for forecasting a wide variety of time series data in many disciplines. This is due to their favorable modeling properties of simplicity, fault and noise tolerance and their capability to adapt to process changes. Nevertheless, applications in climatology have been less widespread than other disciplines such as economics. In this study, feed-forward neural-network (FFNN) and autoregression (AR) time series models are used in forecasting the annual air temperature time series data in Jordan. The performance of the two predictors was compared using out-of-sample forecasts. The test period was shifted through the whole available time. As demonstrated by the forecasting experiments, the FFNN models gave better forecasts and were able to identify the dynamics of the temperature time series and gave more realistic forecasts. Both predictors showed a cooling trend in annual air temperatures for the coming 10 years.

KW - Air temperature

KW - Autoregression

KW - Forecasting

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=34548182281&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548182281&partnerID=8YFLogxK

M3 - Article

VL - 3

SP - 44

EP - 48

JO - Journal of Mathematics and Statistics

JF - Journal of Mathematics and Statistics

SN - 1549-3644

IS - 2

ER -