scholarly journals The Wisdom of the Data: Getting the Most Out of Univariate Time Series Forecasting

Forecasting ◽  
2021 ◽  
Vol 3 (3) ◽  
pp. 478-497
Author(s):  
Fotios Petropoulos ◽  
Evangelos Spiliotis
Keyword(s):  
Time Series ◽  
Computational Cost ◽  
Forecast Accuracy ◽  
Time Series Forecasting ◽  
Data Manipulation ◽  
Combining Forecasts ◽  
Univariate Time Series ◽  
Forecasting Performance ◽  
Wisdom Of The Crowds ◽  
Theta Method

Forecasting is a challenging task that typically requires making assumptions about the observed data but also the future conditions. Inevitably, any forecasting process will result in some degree of inaccuracy. The forecasting performance will further deteriorate as the uncertainty increases. In this article, we focus on univariate time series forecasting and we review five approaches that one can use to enhance the performance of standard extrapolation methods. Much has been written about the “wisdom of the crowds” and how collective opinions will outperform individual ones. We present the concept of the “wisdom of the data” and how data manipulation can result in information extraction which, in turn, translates to improved forecast accuracy by aggregating (combining) forecasts computed on different perspectives of the same data. We describe and discuss approaches that are based on the manipulation of local curvatures (theta method), temporal aggregation, bootstrapping, sub-seasonal and incomplete time series. We compare these approaches with regards to how they extract information from the data, their computational cost, and their performance.

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

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.

scholarly journals Evaluation of the Transformer Architecture for Univariate Time Series Forecasting

2021 ◽  
pp. 106-115
Author(s):  
Pedro Lara-Benítez ◽  
Luis Gallego-Ledesma ◽  
Manuel Carranza-García ◽  
José M. Luna-Romera
Keyword(s):  
Time Series ◽  
Time Series Forecasting ◽  
Univariate Time Series

scholarly journals Explaining Bad Forecasts in Global Time Series Models

2021 ◽  
Vol 11 (19) ◽  
pp. 9243
Author(s):  
Jože Rožanec ◽  
Elena Trajkova ◽  
Klemen Kenda ◽  
Blaž Fortuna ◽  
Dunja Mladenić
Keyword(s):  
Time Series ◽  
Model Performance ◽  
Time Series Forecasting ◽  
Modular Architecture ◽  
Detection Algorithms ◽  
Training Samples ◽  
Forecasting Performance ◽  
Global Time ◽  
Real World Datasets ◽  
Value Changes

While increasing empirical evidence suggests that global time series forecasting models can achieve better forecasting performance than local ones, there is a research void regarding when and why the global models fail to provide a good forecast. This paper uses anomaly detection algorithms and explainable artificial intelligence (XAI) to answer when and why a forecast should not be trusted. To address this issue, a dashboard was built to inform the user regarding (i) the relevance of the features for that particular forecast, (ii) which training samples most likely influenced the forecast outcome, (iii) why the forecast is considered an outlier, and (iv) provide a range of counterfactual examples to understand how value changes in the feature vector can lead to a different outcome. Moreover, a modular architecture and a methodology were developed to iteratively remove noisy data instances from the train set, to enhance the overall global time series forecasting model performance. Finally, to test the effectiveness of the proposed approach, it was validated on two publicly available real-world datasets.

scholarly journals A COMPARATIVE ANALYSIS OF ALTERNATIVE UNIVARIATE TIME SERIES MODELS IN FORECASTING TURKISH INFLATION

2012 ◽  
Vol 13 (2) ◽  
pp. 275-293 ◽  
Author(s):  
A. Nazif Çatık ◽  
Mehmet Karaçuka
Keyword(s):  
Time Series ◽  
Time Series Models ◽  
Unobserved Components ◽  
Inflation Forecasting ◽  
Long Run ◽  
Univariate Time Series ◽  
Forecasting Performance ◽  
Unobserved Components Model ◽  
One Step ◽  
Conventional Models

This paper analyses inflation forecasting power of artificial neural networks with alternative univariate time series models for Turkey. The forecasting accuracy of the models is compared in terms of both static and dynamic forecasts for the period between 1982:1 and 2009:12. We find that at earlier forecast horizons conventional models, especially ARFIMA and ARIMA, provide better one-step ahead forecasting performance. However, unobserved components model turns out to be the best performer in terms of dynamic forecasts. The superiority of the unobserved components model suggests that inflation in Turkey has time varying pattern and conventional models are not able to track underlying trend of inflation in the long run.

scholarly journals Combining Qualitative and Time Series Forecasting to Increase the Forecasting Accuracy for Instant Noodle Sales in Thailand

2016 ◽  
Vol 3 (3) ◽  
pp. 1
Author(s):  
Teerada Khamphinit ◽  
Pornthipa Ongkunaruk
Keyword(s):  
Time Series ◽  
Demand Forecasting ◽  
Forecast Accuracy ◽  
Time Series Forecasting ◽  
Forecasting Accuracy ◽  
Sales Data ◽  
Average Percentage ◽  
Sales Forecast ◽  
Instant Noodle ◽  
Instant Noodles

<p>Demand forecasting is very important for the planning process. The forecast accuracy affects the efficiency of the procurement, production and delivery processes. Our research has the objective of increasing the sales forecasting accuracy of instant noodles for a case study company in Thailand. Many factors affect the sales of instant noodles, such as promotion, other commodities’ prices, national disaster and production capacity. Thus, we collected historical monthly sales data, analysed the data and their pattern and considered whether the data were irregular due to those factors. After obtaining the forecast data, data intervention by adjustment of the irregular effects was performed in accordance with our experience and judgement. Next, we used the predictor function in the Crystal Ball software to determine the best time series forecasting method for actual and adjusted sales data. Then, we verified the result with the actual sales data for one year. The result showed that the adjustment could increase the sales forecast accuracy by 46.14%, 22.53% and 56.42% for products A, B and C, respectively. In summary, the mean average percentage sales forecast error after adjustment was 6.48%–11.62%, which is better than the current method of forecasting based on experts.  </p><p><strong>Keywords</strong>: Instant Noodle; Intervention; Qualitative Forecasting; Sales Adjustment; Time Ser ies Forecasting </p>

Sign in / Sign up

Export Citation Format

Share Document