The Effect of Differencing in Stationary and Accuracy of Time Series in Predicting of Lake Level

Document Type : Complete scientific research article



Background and Objectives: One of the most important assumptions in the modeling of time series, it is to be stationary. The amount of stationary can be various, so that different definitions exist such as first order and second order stationary and strong and strict stationary. Therefore, this study cover the effect of differencing on the stationary value as well as the precision of the ARMA, ARIMA and SARIMA models in the modeling and monthly prediction of time series.
Materials and Methods: For this purpose, 96 years data of lake level, which is monthly measurement related to Michigan-Huron‌‌ Lake on the border of United States and Canada, are used. The 76-years of primary utilized for training and the rest of 20-years are used for validation. Firstly, the existing of the trend and period components in the time-series were assessed using Fischer and man-Kendal tests. These two components are the main factors in the appearance of non-stationary in time series. Therefore seasonal differencing, non-seasonal differencing and both of them at same time were measured and their results were compared by non- differencing data. To assessment of achieved time-series differencing, the ACF diagram and generalized Dicky-Fouler test were utilized. The type and amount of required parameters in different models were determined by ACF diagram. Then, each of series was modeled and predicted using appropriate model. The results indicated that there is not a certain trend and period in series. However, the using of seasonal differencing increased the stationary but non-seasonal differencing lead to non-stationary of these time series. The most increasing in stationary was indicated by using of seasonal and non-seasonal differencing. Due to ACF diagram, using both of differencing results in use of seasonal parameters in model. Therefore, series without differencing with ARMA model and series with seasonal differencing with SARIMA are modeled.
Results: The investigations showed the concurrent using of seasonal and non-seasonal differencing has the most impact on the rate of getting stationary alignment of the Lake in compare with other methods. As a result, the numbers of model needed to achieve the most accurate predictions were reduced in large scale. In such a way in non-differencing situation, 1444 model of ARMA were needed that this amount in situation of seasonal differencing and non-seasonal differencing were reduced in 64 models of SARIMA. On the other hand, by reducing much more number of parameters (two parameters) in SARIMA model, similar result is even better than ARMA model with 21 parameters.
Conclusion: The results showed that the more making stationary of monthly lake level which itself is stationary, reduces the number of models and the number of model's parameters needed to achieve the best outcome too much. For this purpose, combined differencing made the series stationary more than the other methods


1.Azad Talatapeh, N., Behmanesh, J., and Montasari, M. 2013. Predicting Potential
Evapotranspiration Using Time Series Models (Case study: Urmia). J. Water Soil.
27: 1. 213-223. (In Persian)
2.Bolyani, Y., Fazelnia, G., and Bayat, A. 2012. Analysis and modeling annual temperature of
Shiraz using ARIMA model. Geographic Space. 12: 38. 127-144. (In Persian)
3.Brockwell, P.J., and Davis, R.A. 1991. Time series: theory and methods. Second edition,
Springer Science & Business Media, NY, 577p.
4.Chebaane, M., Salas, J.D., and Boes, D.C. 1995. Product periodic autoregressive processes for
modeling intermittent monthly stream flows. J. Water Resour. Res. 31: 6. 1513-1518.
5.Çimen, M., and Kisi, O. 2009. Comparison of two different data-driven techniques in
modeling lake level fluctuations in Turkey. J. Hydrol. 378: 3-4. 253-262.
6.Cryer, J.D., and Chan, K.S. 2008. Time Series Analysis With Applications in R. Second Ed.,
Springer, NY, 491p.
7.Giri, A., and Singh, N.B. 2014. Comparison of Artificial Neural Network Algorithm for Water
Quality Prediction of River Ganga. Environ. Res. J. 8: 2. 55-63.
8.Hirsch, R.M., and Slack, J.R. 1984. A nonparametric trend test for seasonal data with serial
dependence. J. Water Resour. Res. 20: 6. 727-732.
9.Kashyap, R.L., and Ramachandra Rao, A. 1976. Dynamic stochastic models from empirical
data. Academic press, NY, 352p.
10.Khatibi, R., Ghorbani, M., Naghipour, L., Jothiprakash, V., Fathima, T., and Fazelifard, M.
2014. Inter-comparison of time series models of lake levels predicted by several modeling
strategies. J. Hydrol. 511: 530-545.
11.Khazaee, M., and Mirzaee, M. 2014. Forecasting the climatic variables using time series
analysis of Zohre catchment. Sci. J. Manage. Syst. 14: 34. 233-250. (In Persian)
12.Kisi, Ö. 2004. River flow modeling using artificial neural networks. J. Hydrol. Engin.
9: 1. 60-63.
13.Kisi, O., and Cigizoglu, H.K. 2007. Comparison of different ANN techniques in river flow
prediction. J. Civil Engin. Environ. Syst. 24: 3. 211-231.
14.Kisi, O., Shiri, J., Karimi, S., Shamshirband, Sh., Motamedi, Sh., Petkovic, D., and Hashim,
R. 2015. A survey of water level fluctuation predicting in Urmia Lake using support vector
machine with firefly algorithm. J. Appl. Math. Comp. 270: 731-743.
15.Makarynska, D., and Makarynskyy, O. 2008. Predicting sea-level variations at the Cocos
(Keeling) Islands with artificial neural networks. J. Comp. Geosci. 34: 12. 1910-1917.
16.Marco, J.B., Harboe, R., and Salas, J.D. 1993. Stochastic hydrology and its use in water
resources systems simulation and optimization. Springer Science & Business Media,
Peniscola, Spain, 483p.
17.Maroofi, S., Khotar, B., Sadeghifar, M., Parsafar, N., and Ildormi, A. 2014. Forecasting the
drought using SARIMA time series and SPI index in the central region of the Hamedan
province. 28: 1. 213-235. (In Persian)
18.Omidi, R., Radmanesh, F., and Zarei, H. 2013. River flow prediction using stochastic
models. The First National Conference on Challenges on Water Resources and Agricultural,
13th February, Khorasgan Branch of Islamic Azad university, Iran, 8p. (In Persian)
19.Peña, D., Tiao, G.C., and Tsay, R.S. 2011. A course in time series analysis. John Wiley &
Sons, INC, NY, 460p.
20.Poormohammadi, S., Malekinezhad, H., and Poorshareyati, R. 2013. Comparison of ANN
and time series appropriately in prediction of ground water table (Case study: Bakhtegan
basin). J. Water Soil Cons. 20: 4. 251-262. (In Persian)
21.Said, S.E., and Dickey, D.A. 1984. Testing for unit roots in autoregressive-moving average
models of unknown order. Biometrika. 71: 3. 599-607.
22.Salas, J.D., Delleur, J.W., Yevjevich, V., and Lane, W.L. 1980. Applied modeling of
hydrologic time series. Water Resources Publication, Colorado, 484p.
23.Shafaei, M., and Kisi, O. 2015. Lake Level Forecasting Using Wavelet-SVR, WaveletANFIS and Wavelet-ARMA Conjunction Models. J. Water Resour. Manage. 30: 1. 79-97.
24.Shamim, M.A., Hassan, M., Ahmad, S., and Zeeshan, M. 2015. A comparison of Artificial
Neural Networks (ANN) and Local Linear Regression (LLR) techniques for predicting
monthly reservoir levels. KSCE J. Civil Engin. 8p. DOI: 10.1007/s12205-015-0298-z.
25.Sharma, N., Zakaullah, M., Tiwari, H., and Kumar, D. 2015. Runoff and sediment yield
modeling using ANN and support vector machines: a case study from Nepal watershed.
Modeling Earth Systems Environment 1 (23), 8p. DOI: 10.1007/s40808-015-0027-0.
26.St-Hilaire, A., Ouarda, T.B., Bargaoui, Z., Daigle, A., and Bilodeau, L. 2012. Daily river
water temperature forecast model with a k-nearest neighbour approach. J. Hydrol. Proc.
26: 9. 1302-1310.
27.Tao, P.C., and Delleur, J.W. 1976. Seasonal and nonseasonal ARMA models in hydrology.
J. Hydrol. Div. 102: 10. 1541-1559.
28.Veisipoor, H., Samakoosh, J.M., Sahneh, B., and Yousofi, Y. 2010. Analysis prediction the
precipitation and temperature using time series models (ARIMA). Geography. 4: 12. 63-70.
(In Persian)
29.Wald, A.B., and Wolfowitz, J.A. 1943. An exact test for randomness Ian the non-parametric
case based on serial correlation. The Annals of Mathematical Statistics. 14: 4. 378-88.
30.Wang, W.C., Chau, K.W., Xu, D.M., and Chen, X.Y. 2015. Improving Forecasting Accuracy
of Annual Runoff Time Series Using ARIMA Based on EEMD Decomposition. J. Water
Resour. Manage. 29: 8. 2655-2675.
31.Wilcox, D.A., Thompson, T.A., Booth, R.K., and Nicholas, J.A. 2007. Water-level
variability and water availability. J. Great Lakes. Geological Survey Circular. 1311, U.S.,
32.Jabbari Gharabagh, S., Rezaei, H., and Mohammadnezhad, B. 2015. Comparison of
reconstructed phase space and chaotic behavior of Nazloochay river flow at different
temporal scales. J. Water Soil Cons. 22: 5. 135-151. (In Persian)
33.Rajaei, T., and Ebrahimi, H. 2015. Application of wavelet-neural network model for
forecasting of groundwater level time series with non-stationary and nonlinear
characteristics. J. Water Soil Cons. 22: 5. 99-115. (In Persian)
34.Ahmadi, F., Dinpazhooh, Y., Fakherifard, A., Khalili, K., and Darbandi, S. 2015. Comparing
Nonlinear Time Series Models and Genetic Programming for Daily River Flow Forecasting
(Case study: Barandouz-Chai River). J. Water Soil Cons. 22: 1. 151-169. (In Persian)
35.Rajaei, T., and Broomand, A. 2016. Prediction of Monthly Dissolved Oxygen Using Wavelet
and Artificial Neural Network Combined Model. J. Water Soil Cons. 22: 6. 153-169.
(In Persian)
36.Altunkaynak, A. 2014. Predicting water level fluctuations in Lake Michigan-Huron using
wavelet-expert system methods. Water resources management. 28: 8. 2293-2314.
37.Coulibaly, P. 2010. Reservoir computing approach to Great Lakes water level forecasting.
J. Hydrol. 381: 1. 76-88.