Yıl 2017, Cilt 9, Sayı 17, Sayfalar 147 - 154 2017-11-27
| | | |

## trenNEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING?Neural Network Data Preprocessing: Is It Necessary For Time Series Forecasting?

#### Hakan PABUÇCU [1]

##### 187 370

Neural networks (NNs) are a commonly used method to solve the time series-forecasting problem. NNs have some advantages compared with traditional forecasting models, such as auto regressive moving average or auto regressive integrated moving average. NNs do not need to have any statistical assumption like normal distribution. However, data preprocessing, normalization, trend adjusting, seasonal adjusting, or both differencing can introduce better results in some studies. In this study, we have tried to investigate whether data preprocessing methods are useful for time series data, which contains trend, seasonality, or unit root. For this purpose, we collected the real time series data belonging to monthly or quarterly figures and used nonlinear autoregressive (NAR) and multilayer perceptron (MLP) models. Although we obtained significant differences between data preprocessing methods, the structure of MLP with differenced variable produced the worst results.

Neural networks (NNs) are a commonly used method to solve the time series-forecasting problem. NNs have some advantages compared with traditional forecasting models, such as auto regressive moving average or auto regressive integrated moving average. NNs do not need to have any statistical assumption like normal distribution. However, data preprocessing, normalization, trend adjusting, seasonal adjusting, or both differencing can introduce better results in some studies. In this study, we have tried to investigate whether data preprocessing methods are useful for time series data, which contains trend, seasonality, or unit root. For this purpose, we collected the real time series data belonging to monthly or quarterly figures and used nonlinear autoregressive (NAR) and multilayer perceptron (MLP) models. Although we obtained significant differences between data preprocessing methods, the structure of MLP with differenced variable produced the worst results.

• Alon, I., Qi, M., & Sadowski, R. J. (2001). Forecasting aggregate retail sales: Journal of Retailing and Consumer Services, 8(3), 147–156. http://doi.org/http://dx.doi.org/10.1016/S0969-6989(00)00011-4
• Bishop, C. M. (1995). Neural networks for pattern recognition. New York, NY: Oxford University Press.
• Findley, D. F., Monsell, B. C., Bell, W. R., Otto, M. C., & Chen, B.-C. (1998). New Capabilities and Methods of the X-12-ARIMA Seasonal-Adjustment Program. Journal of Business & Economic Statistics, 16(2), 127–152. http://doi.org/10.1080/07350015.1998.10524743
• Ghysels, E., Granger, C. W. J., & Siklos, P. L. (1996). Is Seasonal Adjustment a Linear or Nonlinear Data-Filtering Process? Journal of Business & Economic Statistics, 14(3), 374–386. http://doi.org/10.1080/07350015.1996.10524663
• Gorr, W. L. (1994). Editorial: Research prospective on neural network forecasting. International Journal of Forecasting, 10(1), 1–4. http://doi.org/http://dx.doi.org/10.1016/0169-2070(94)90044-2
• Hamzaçebi, C. (2008). Improving artificial neural networks’ performance in seasonal time series forecasting. Information Sciences, 178(23), 4550–4559. http://doi.org/http://dx.doi.org/10.1016/j.ins.2008.07.024
• Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359–366. http://doi.org/http://dx.doi.org/10.1016/0893-6080(89)90020-8
• Hylleberg, S. (1992). Modelling seasonality (1st ed.). Oxford: Oxford University Press.
• Makridakis, S., Andersen, A., Carbone, R., Fildes, R., Hibon, M., Lewandowski, R., … Winkler, R. (1982). The accuracy of extrapolation (time series) methods: Results of a forecasting competition. Journal of Forecasting, 1(2), 111–153. http://doi.org/10.1002/for.3980010202
• Nelson, M., Hill, T., Remus, W., & O’Connor, M. (1999). Time series forecasting using neural networks: should the data be deseasonalized first? Journal of Forecasting, 18(5), 359–367. http://doi.org/10.1002/(SICI)1099-131X(199909)18:5<359::AID-FOR746>3.0.CO;2-P
• Öztemel, E. (2006). Yapay Sinir Ağları (2nd ed.). İatanbul: Papatya Yayıncılık.
• Ripley, B. D. (1996). Pattern recognition and neural networks. http://doi.org/http://dx.doi.org/10.1017/cbo9780511812651
• Sharda, R., & Patil, R. B. (1992). Connectionist approach to time series prediction: an empirical test. Journal of Intelligent Manufacturing, 3(5), 317–323. http://doi.org/10.1007/BF01577272
• Yegnanarayana, B. (2005). Artificial Neural Networks (11th ed.). New Delhi: Prentice-Hall of lndia Private Limited.
• Zhang, G. P., & Kline, D. M. (2007). Quarterly time-series forecasting with neural networks. IEEE Transactions on Neural Networks, 18(6), 1800–1814. http://doi.org/10.1109/TNN.2007.896859
• Zhang, G. P., & Qi, M. (2005). Neural network forecasting for seasonal and trend time series. European Journal of Operational Research, 160, 501–514. http://doi.org/10.1016/j.ejor.2003.08.037
Konular Sosyal MAKALELER Yazar: Hakan PABUÇCU (Sorumlu Yazar)Kurum: BAYBURT ÜNİVERSİTESİÜlke: Turkey
 Bibtex @araştırma makalesi { kilisiibfakademik329247, journal = {Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD)}, issn = {1309-3762}, eissn = {2149-1585}, address = {Kilis 7 Aralık Üniversitesi}, year = {2017}, volume = {9}, pages = {147 - 154}, doi = {10.20990/kilisiibfakademik.329247}, title = {NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING?}, key = {cite}, author = {PABUÇCU, Hakan} } APA PABUÇCU, H . (2017). NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING?. Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD), 9 (17), 147-154. DOI: 10.20990/kilisiibfakademik.329247 MLA PABUÇCU, H . "NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING?". Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD) 9 (2017): 147-154 Chicago PABUÇCU, H . "NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING?". Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD) 9 (2017): 147-154 RIS TY - JOUR T1 - NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING? AU - Hakan PABUÇCU Y1 - 2017 PY - 2017 N1 - doi: 10.20990/kilisiibfakademik.329247 DO - 10.20990/kilisiibfakademik.329247 T2 - Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD) JF - Journal JO - JOR SP - 147 EP - 154 VL - 9 IS - 17 SN - 1309-3762-2149-1585 M3 - doi: 10.20990/kilisiibfakademik.329247 UR - http://dx.doi.org/10.20990/kilisiibfakademik.329247 Y2 - 2017 ER - EndNote %0 Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD) NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING? %A Hakan PABUÇCU %T NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING? %D 2017 %J Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD) %P 1309-3762-2149-1585 %V 9 %N 17 %R doi: 10.20990/kilisiibfakademik.329247 %U 10.20990/kilisiibfakademik.329247 ISNAD PABUÇCU, Hakan . "NEURAL NETWORK DATA PREPROCESSING: IS IT NECESSARY FOR TIME SERIES FORECASTING?". Akademik Araştırmalar ve Çalışmalar Dergisi (AKAD) 9 / 17 (Kasım 2017): 147-154. http://dx.doi.org/10.20990/kilisiibfakademik.329247