COMPARING THE PREDICTION ACCURACY OF LSTM AND ARIMA MODELS FOR TIME-SERIES WITH PERMANENT FLUCTUATION

In developing countries with an unstable economic system, permanent fluctuation in historical data is always a concern. Recognizing dependency and independency of variables are vague and proceeding a reliable forecast model is more complex than other countries. Although linearization of nonlinear multivariate economic time-series to predict, may give a result, the nature of data which shows irregularities in the economic system, should be ignored. New approaches of artificial neural network (ANN) help to make a prediction model with keeping data attributes. In this paper, we used the Tehran Stock Exchange (TSE) intraday data in 10 years to forecast the next 2 months. Long Short-Term Memory (LSTM) from ANN chooses and outputs compared with the autoregressive integrated moving average (ARIMA) model. The results show, although, in long term prediction, the forecast accuracy of both models reduce, LSTM outperforms ARIMA, in terms of error of accuracy, significantly.

Download Full-text

An improved ARIMA model for hydrological simulations

Nonlinear Processes in Geophysics Discussions ◽

10.5194/npgd-1-841-2014 ◽

2014 ◽

Vol 1 (1) ◽

pp. 841-876 ◽

Cited By ~ 2

Author(s):

H. R. Wang ◽

C. Wang ◽

X. Lin ◽

J. Kang

Keyword(s):

Time Series ◽

Time Series Data ◽

Moving Average ◽

Arima Model ◽

Forecast Accuracy ◽

Series Data ◽

Monthly Data ◽

Annual Variations ◽

Characteristic Quantity ◽

Improved Model

Abstract. Auto Regressive Integrated Moving Average (ARIMA) model is often used to calculate time series data formed by inter-annual variations of monthly data. However, the influence brought about by inter-monthly variations within each year is ignored. Based on the monthly data classified by clustering analysis, the characteristics of time series data are extracted. An improved ARIMA model is developed accounting for both the inter-annual and inter-monthly variation. The correlation between characteristic quantity and monthly data within each year is constructed by regression analysis first. The model can be used for predicting characteristic quantity followed by the stationary treatment for characteristic quantity time series by difference. A case study is conducted to predict the precipitation in Lanzhou precipitation station, China, using the model, and the results show that the accuracy of the improved model is significantly higher than the seasonal model, with the mean residual achieving 9.41 mm and the forecast accuracy increasing by 21%.

Download Full-text

A Note On Earnings Forecast Source Superiority

Journal of Applied Business Research (JABR) ◽

10.19030/jabr.v19i3.2174 ◽

2011 ◽

Vol 19 (3) ◽

Author(s):

Donald P. Pagach ◽

Barbara A. Chaney ◽

Bruce C. Branson

Keyword(s):

Small Firms ◽

Moving Average ◽

Arima Model ◽

Decision Rules ◽

Forecast Accuracy ◽

Forecast Model ◽

Earnings Forecast ◽

Font Size ◽

Ex Ante ◽

Marginal Improvement

We examine the forecast accuracy of Value Line analysts relative to the Brown-Rozeff (100)X(011)4 ARIMA model.  We find that for a surprising percentage (35-41%) of our sample of small firms that time series-based earnings per share predictions are more accurate than those obtained from The Value Line Investment Survey.  Further, we document exploitable characteristics of each subgroup that are associated with forecast origin.  In those instances where the seasonal, univariate earnings forecast model identified by Brown and Rozeff (1979) produces more accurate forecasts than Value Line, we find significant differences in firm size, degree of diversification, magnitudes of the autoregressive and seasonal moving-average parameters, residual standard errors, and magnitude of the Ljung-Box Q-statistic.  We use probit regressions to identify ex ante those firms likely to be accurately forecast by each source.  We achieve a marginal improvement in forecast accuracy, which suggests there is potential for using ex ante decision rules to improve forecast accuracy.

Download Full-text

Time Series Features Extraction and Forecast from Multi-feature Stocks with Hybrid Deep Neural Networks.

Recent Advances in Computer Science and Communications ◽

10.2174/2666255813999200625220302 ◽

2020 ◽

Vol 13 ◽

Author(s):

Ye Xu ◽

Xun Yuan

Keyword(s):

Neural Networks ◽

Time Series ◽

Hybrid Model ◽

Deep Neural Networks ◽

Arima Model ◽

Forecast Accuracy ◽

Forecast Model ◽

Closing Price ◽

Massive Information ◽

Better Than

Background: Forecasting of time series stock data is important in financial related works. Stock data usually have multifeatures such as opening price, closing price and so on. The traditional forecast methods, however, is mainly applied to one feature – closing price, or a few, like four or five features. The massive information hidden in the multi-feature data is not thoroughly discovered and used. Objective: Find a method to make used of all information of multi-features and get a forecast model. Method: LSTM based models are introduced in this paper. For comparison, three models are used and they are single LSTM model, hybrid model of LSTM-CNN, and traditional ARIMA model. Results: Experiments with different models are performed on stock data with 50 and 230 features, respectively. Results show that MSE of single LSTM model is 2.4% lower than ARIMA model and MSE of LSTM-CNN model is 12.57% lower than that of single LSTM model on 50 features data. On 230 features data, LSTM-CNN model is found to be improved by 23.41% in forecast accuracy. Conclusion: In this paper, we use three different models – ARIMA, single LSTM and LSTM-CNN hybrid model – to forecast rise and fall of multi-features stock data. It’s found that single LSTM model is better than traditional ARIMA model on the average, and LSTM-CNN hybrid model is better than single LSTM model on 50-feature stock data. What’s more, we use LSTM-CNN model to perform experiments on stock data with 50 and 230 features, respectively. And is found that results of the same model on 230 features data is better than that on 50 features data. It’s proved in our work that the LSTM-CNN hybrid model is better than other models and experiments on stock data with more features could result in better outcomes. We’ll do more works on hybrid models next.

Download Full-text

Predicting Closed Price Time Series Data Using ARIMA Model

Modern Applied Science ◽

10.5539/mas.v12n11p181 ◽

2018 ◽

Vol 12 (11) ◽

pp. 181 ◽

Cited By ~ 5

Author(s):

S. AL Wadi ◽

Mohammad Almasarweh ◽

Ahmed Atallah Alsaraireh

Keyword(s):

Time Series ◽

Time Series Data ◽

Stock Exchange ◽

Moving Average ◽

Arima Model ◽

Series Data ◽

Price Forecasting ◽

Forecasting Accuracy ◽

Autoregressive Integrated Moving Average ◽

Short Term Prediction

Closed price forecasting plays a main rule in finance and economics which has encouraged the researchers to introduce a fit model in forecasting accuracy. The autoregressive integrated moving average (ARIMA) model has developed and implemented in many applications. Therefore, in this article the researchers utilize ARIMA model in predicting the closed time series data which have been collected from Amman Stock Exchange (ASE) from Jan. 2010 to Jan. 2018. As a result this article shows that the ARIMA model has significant results for short-term prediction. Therefore, these results will be helpful for the investments.

Download Full-text

Predictive Efficiency Comparison of ARIMA-Time-Series and the Grey System GM(1,1) Forecast Model on Forecasting the Incidence Rate of Hepatitis B

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.709.836 ◽

2013 ◽

Vol 709 ◽

pp. 836-839 ◽

Cited By ~ 4

Author(s):

Yin Ping Chen ◽

Ai Ping Wu ◽

Cui Ling Wang ◽

Hai Ying Zhou ◽

Si Zhao

Keyword(s):

Time Series ◽

Hepatitis B ◽

Incidence Rate ◽

Moving Average ◽

Arima Model ◽

Forecast Model ◽

Grey System ◽

Gray System ◽

Efficiency Comparison ◽

White Noise Sequence

To compare the stochastic autoregressive integrated moving average (ARIMA) model and the grey system GM(1,1) model to predict the hepatitis B incidence in Qianan. Considering the Box-Jenkins modeling and GM(1,1) model approach, hepatitis B incidence was collected monthly from 2004 to 2011, a SARIMA model and a gray system GM(1,1) model were fit. Then, these models were used for calculating hepatitis B incidence for the last 6 observations compared with observed data. The constructed models were performed to predict the monthly incidence rate in 2013. The model SARIMA(0,1,1)(0,1,1)12 and was established finally and the residual sequence was a white noise sequence. Using Excel 2003 to establish the gray system GM(1,1) model of hepatitis B incidence and evaluating the accuracy of the mode as well as forecasting. By posterior-error-test (C=0.435, p=0.821) and residual test, the model accuracy was qualified. It was necessary and practical to apply the approach of ARIMA model in fitting time series to predict hepatitis within a short lead time. The prediction results showed that the hepatitis B incidence in 2013 had a slight upward trend.

Download Full-text

A COMPARATIVE STUDY BETWEEN UNIVARIATE AND BIVARIATE TIME SERIES MODELS FOR CRUDE PALM OIL INDUSTRY IN PENINSULAR MALAYSIA

MALAYSIAN JOURNAL OF COMPUTING ◽

10.24191/mjoc.v5i1.6760 ◽

2020 ◽

Vol 5 (1) ◽

pp. 374

Author(s):

Pauline Jin Wee Mah ◽

Nur Nadhirah Nanyan

Keyword(s):

Time Series ◽

Palm Oil ◽

Moving Average ◽

Forecast Accuracy ◽

Peninsular Malaysia ◽

Time Series Models ◽

Crude Palm Oil ◽

Univariate Time Series ◽

Import And Export ◽

Bivariate Time Series

The main purpose of this study is to compare the performances of univariate and bivariate models on four time series variables of the crude palm oil industry in Peninsular Malaysia. The monthly data for the four variables, which are the crude palm oil production, price, import and export, were obtained from Malaysian Palm Oil Board (MPOB) and Malaysian Palm Oil Council (MPOC). In the first part of this study, univariate time series models, namely, the autoregressive integrated moving average (ARIMA), fractionally integrated autoregressive moving average (ARFIMA) and autoregressive autoregressive (ARAR) algorithm were used for modelling and forecasting purposes. Subsequently, the dependence between any two of the four variables were checked using the residuals’ sample cross correlation functions before modelling the bivariate time series. In order to model the bivariate time series and make prediction, the transfer function models were used. The forecast accuracy criteria used to evaluate the performances of the models were the mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The results of the univariate time series showed that the best model for predicting the production was ARIMA while the ARAR algorithm were the best forecast models for predicting both the import and export of crude palm oil. However, ARIMA appeared to be the best forecast model for price based on the MAE and MAPE values while ARFIMA emerged the best model based on the RMSE value. When considering bivariate time series models, the production was dependent on import while the export was dependent on either price or import. The results showed that the bivariate models had better performance compared to the univariate models for production and export of crude palm oil based on the forecast accuracy criteria used.

Download Full-text

Forecasting Nigerian Stock Exchange Returns: Evidence from Autoregressive Integrated Moving Average (ARIMA) Model

SSRN Electronic Journal ◽

10.2139/ssrn.1633006 ◽

2010 ◽

Author(s):

Emenike Kalu O.

Keyword(s):

Stock Exchange ◽

Moving Average ◽

Arima Model ◽

Autoregressive Integrated Moving Average

Download Full-text

Identification of monthly municipal water demand system based on autoregressive integrated moving average model tuned by particle swarm optimization

Journal of Hydroinformatics ◽

10.2166/hydro.2017.035 ◽

2017 ◽

Vol 19 (2) ◽

pp. 261-281 ◽

Cited By ~ 7

Author(s):

Sahbi Boubaker

Keyword(s):

Time Series ◽

Particle Swarm Optimization ◽

Water Demand ◽

Moving Average ◽

Arima Model ◽

Particle Swarm ◽

Demand System ◽

Swarm Optimization ◽

Municipal Water ◽

Average Absolute Relative Error

In this paper, a modeling-identification approach for the monthly municipal water demand system in Hail region, Saudi Arabia, is developed. This approach is based on an auto-regressive integrated moving average (ARIMA) model tuned by the particle swarm optimization (PSO). The ARIMA (p, d, q) modeling requires estimation of the integer orders p and q of the AR and MA parts; and the real coefficients of the model. More than being simple, easy to implement and effective, the PSO-ARIMA model does not require data pre-processing (original time-series normalization for artificial neural network (ANN) or data stationarization for traditional stochastic time-series (STS)). Moreover, its performance indicators such as the mean absolute percentage error (MAPE), coefficient of determination (R2), root mean squared error (RMSE) and average absolute relative error (AARE) are compared with those of ANN and STS. The obtained results show that the PSO-ARIMA outperforms the ANN and STS approaches since it can optimize simultaneously integer and real parameters and provides better accuracy in terms of MAPE (5.2832%), R2 (0.9375), RMSE (2.2111 × 105m3) and AARE (5.2911%). The PSO-ARIMA model has been implemented using 69 records (for both training and testing). The results can help local water decision makers to better manage the current water resources and to plan extensions in response to the increasing need.

Download Full-text

Towards optimal ATM cash replenishment using time series analysis

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-201953 ◽

2021 ◽

pp. 1-13

Author(s):

Muhammad Rafi ◽

Mohammad Taha Wahab ◽

Muhammad Bilal Khan ◽

Hani Raza

Keyword(s):

Time Series ◽

Moving Average ◽

Arima Model ◽

Mean Absolute Percentage Error ◽

Percentage Error ◽

Absolute Percentage Error ◽

Automatic Teller Machine ◽

Proposed Model ◽

Auto Regressive ◽

Specific Amount

Automatic Teller Machine (ATM) are still largely used to dispense cash to the customers. ATM cash replenishment is a process of refilling ATM machine with a specific amount of cash. Due to vacillating users demands and seasonal patterns, it is a very challenging problem for the financial institutions to keep the optimal amount of cash for each ATM. In this paper, we present a time series model based on Auto Regressive Integrated Moving Average (ARIMA) technique called Time Series ARIMA Model for ATM (TASM4ATM). This study used ATM back-end refilling historical data from 6 different financial organizations in Pakistan. There are 2040 distinct ATMs and 18 month of replenishment data from these ATMs are used to train the proposed model. The model is compared with the state-of- the-art models like Recurrent Neural Network (RNN) and Amazon’s DeepAR model. Two approaches are used for forecasting (i) Single ATM and (ii) clusters of ATMs (In which ATMs are clustered with similar cash-demands). The Mean Absolute Percentage Error (MAPE) and Symmetric Mean Absolute Percentage Error (SMAPE) are used to evaluate the models. The suggested model produces far better forecasting as compared to the models in comparison and produced an average of 7.86/7.99 values for MAPE/SMAPE errors on individual ATMs and average of 6.57/6.64 values for MAPE/SMAPE errors on clusters of ATMs.

Download Full-text

Research on GDP Forecast Analysis Combining BP Neural Network and ARIMA Model

Computational Intelligence and Neuroscience ◽

10.1155/2021/1026978 ◽

2021 ◽

Vol 2021 ◽

pp. 1-10

Author(s):

Shaobo Lu

Keyword(s):

Neural Network ◽

Time Series ◽

Prediction Model ◽

Error Correction ◽

Relative Error ◽

Bp Neural Network ◽

Arima Model ◽

Forecast Model ◽

Price Prediction ◽

The Neural Network

Based on the BP neural network and the ARIMA model, this paper predicts the nonlinear residual of GDP and adds the predicted values of the two models to obtain the final predicted value of the model. First, the focus is on the ARMA model in the univariate time series. However, in real life, forecasts are often affected by many factors, so the following introduces the ARIMAX model in the multivariate time series. In the prediction process, the network structure and various parameters of the neural network are not given in a systematic way, so the operation of the neural network is affected by many factors. Each forecasting method has its scope of application and also has its own weaknesses caused by the characteristics of its own model. Secondly, this paper proposes an effective combination method according to the GDP characteristics and builds an improved algorithm BP neural network price prediction model, the research on the combination of GDP prediction model is currently mostly focused on the weighted form, and this article proposes another combination, namely, error correction. According to the price characteristics, we determine the appropriate number of hidden layer nodes and build a BP neural network price prediction model based on the improved algorithm. Validation of examples shows that the error-corrected GDP forecast model is also better than the weighted GDP forecast model, which shows that error correction is also a better combination of forecasting methods. The forecast results of BP neural network have lower errors and monthly prices. The relative error of prediction is about 2.5%. Through comparison with the prediction results of the ARIMA model, in the daily price prediction, the relative error of the BP neural network prediction is 1.5%, which is lower than the relative error of the ARIMA model of 2%.

Download Full-text