River flow FORECASTING for a region has a special and important role for optimal allocation of water resources. In this research, for FORECASTING river flow process, Fuzzy Inference System (FIS) is used. Three parameters including precipitation, temperature and daily discharge are used for FORECASTING of daily river flow of Lighvan River located in Lighvanchai watershed. For the initial preprocessing, the randomness of data was examined by return points test. Then, for determination of the optimum lags for input parameters, correlogram of data was considered. Finally to investigate the effects of temperature on river flow FORECASTING, the process were done for any months separately. Assessments of prediction by using various criteria such as Nash-Sutcliff coefficient showed that FIS model had high precision (CNS=0.9976) and low error (RMSE=0.0113) in prediction which shows that the FIS model can be employed successfully in river flow FORECASTING. Final assessment of the results was also revealed the effects of temperature on prediction in some months (April and December).

Main goal of this paper is Investigation and comparative between Accounting model information content with economic model for operating cash flow in Tehran exchange.Result shows that meaningful relation between CVA, Net profit with operating cash flow. Also result paper show that for operating cash flow FORECASTING have equal information content based on two models.

In this paper the PERFORMANCE of iterated and direct autoregressive models in FORECASTING Iranian inflation has been evaluated in horizons 1, 2, 3 and 4 steps ahead. The results show that the forecast accuracy of direct method compared to iterated method depends on the information criteria. In FORECASTING literature, lag selection is done as cumulative. This paper also investigate whether the use of all possible combination of lags, rather than using cumulative lags can lead to improve forecast accuracy. Our findings show that the optimal combination of lags changes depending on forecast horizon, so that the best combination of lags in the horizon 1 and 2 is the first lag, and in the horizon 3 and 4, are the first and fourth lags. Also using IC method to reduce systematic error does not improve forecast accuracy.

This paper discusses the problems of short-term FORECASTING of cryptocurrency time series using a supervised machine learning (ML) approach. For this goal, we applied two of the most powerful ensemble methods including Random Forests (RF) and Stochastic Gradient Boosting Machine (SGBM). As the dataset was collected from daily close prices of three of the most capitalized coins: Bitcoin (BTC), Ethereum (ETH) and Ripple (XRP), and as features we used  past price information and technical indicators (moving average). To check the effectiveness of these models we made an out-of-sample forecast for selected time series by using the one step ahead technique. The accuracy rate of the forecasted prices by using RF and GBM were calculated. The results verify the applicability of the ML ensembles approach for the FORECASTING of cryptocurrency prices. The out of sample accuracy of short-term prediction daily close prices obtained by the SGBM and RF in terms of Mean Absolut Percentage Error (MAPE) for the three most capitalized cryptocurrencies (BTC, ETH, and XRP) were within 0.92-2.61 %.

In this paper, we evaluate the PERFORMANCE of two machine learning architectures— Recurrent Neural Networks (RNN) and Transformer-based models—on four commodity-based company indices from the Tehran Stock Exchange. The Transformer-based models used in this study include AutoFormer, FEDformer, Informer, and PatchTST, while the RNN-based models consist of GRU and LSTM. The dataset comprises daily observations collected from April 20, 2020, to November 20, 2024. To enhance the generalization power of the models and prevent overfitting, we employ two techniques: splitting the training and test samples, and applying regularization methods such as dropout. Hyperparameters for all models were selected using a visual method. Our results indicate that the PatchTST model outperforms other methods in terms of Root Mean Squared Error (RMSE) for both 1-day and 5-day (1-week) FORECASTING horizons. The FEDformer model also demonstrates promising PERFORMANCE, particularly for FORECASTING the MetalOre time series. In contrast, the AutoFormer model performs relatively poorly for longer FORECASTING horizons, while the GRU and LSTM models yield mixed results. These findings underscore the significant impact of model selection and FORECASTING horizon on the accuracy of time series forecasts, emphasizing the importance of careful model choice and hyperparameter tuning for achieving optimal PERFORMANCE.