FCSTATS: Stata module to compute time series forecast accuracy statistics
1
Citation
0
Reference
20
Related Paper
Citation Trend
Abstract:
fcstats calculates several measures of forecast accuracy for one or two forecast series. The measures include root mean squared error (RMSE), mean absolute error (MAE), mean absolute percent error (MAPE) and Theil's U.Keywords:
Mean absolute error
Cite
The availability of short-term forecast weather model for a particular country or region is essential for operation planning of energy systems. This paper presents the first step by a group of researchers at UAE University to establish a weather model for the UAE using the weather data for at least 10 years and employing various models such as classical empirical models, artificial neural network (ANN) models, and time-series regression models with autoregressive integrated moving-average (ARIMA). This work uses time-series regression with ARIMA modeling to establish a model for the mean daily and monthly global solar radiation (GSR) for the city of Al-Ain, United Arab Emirates. Time-series analysis of solar radiation has shown to yield accurate average long-term prediction performance of solar radiation in Al-Ain. The model was built using data for 10 years (1995–2004) and was validated using data of three years (2005–2007), yielding deterministic coefficients (R 2
Mean absolute error
Predictive modelling
Cite
Citations (47)
Mean absolute error
Univariate
Absolute deviation
Cite
Citations (2)
Nobody can accurately predict amount of sales revenue expected.There is need for accuracy to eliminate risk of revenue loss.Surveys have shown that accuracy is the most important criterion in selecting a forecasting strategy.Which criterion then provides the most accurate forecast?Human judgment and software have been used to address this issue but the question is how do we judge the accuracy?Is there revenue leakage due to over or under forecasting, resulting from lack of accuracy on the methods used?This paper evaluates accuracy of revenue forecasted over eleven years in a semi-autonomous Tax administration, a revenue collection body which collects taxes on behalf of government in Malawi.We investigate whether forecasting method which was used was the most accurate, unbiased and efficient, and whether better forecasts could have been used.We compare different Mean Absolute Deviation (MAD) and Mean Square Error(MSE) calculated from different time series methods using software CB predictor to establish the most accurate method for forecasting revenue collection for this organization.Actual data provided by Ministry of Finance in Malawi was used.The study shows that the forecasting method used by the organization was not most accurate than using the Last Value Forecasting Method which had the lowest MAD among the other forecasting time series method that were tested.The current method used by the organization had huge forecasting errors.The paper further recommends the use of Last Value Forecasting Method, and suggests other forecasting objectives the organization's may have other than accuracy.
Tax administration
Cite
Citations (0)
This study is focused on predicting the consumption of Petroleum (Thousands of Barrels per year) in Nigeria. Autoregressive integrated moving average (ARIMA), Linear Regression (LR) and Random Forest Regression (RFR) models were fitted to predict the consumption of Petroleum. The prediction accuracy of these models was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Coefficient of determination (R^2 ) metrics. The Petroleum dataset spanned a period of 37 years (1980-2017) and it was spilted into train and test at the ratio of 70:30 respectively to reduce overfitting. The result obtained revealed that the two machine learning models: LR and RFR outperformed the ARIMA model with lower values of prediction accuracy in terms of MAE, MAPE, RMSE and .
Overfitting
Mean absolute error
Cite
Citations (0)
A study was conducted on time-series data on rice production in India. Box-Jenkins Autoregressive Integrated Moving Average (ARIMA) time-series process was considered for predicting country's rice production using the time series data from 1950–51 to 2017–18. Data from 1950–51 to 2014–15 were used for model development and three years data from 2015–16 and 2017–18 were kept for validation The augmented Dicky Fuller test was applied to test stationarity in data set. Root mean square error. Based on ACF and PACF, the model was defined and tested for its suitability. Akaike information criterion and Bayesian information criterion were used to judge the suitability of the model to be fitted. The performance of the fitted model was examined using mean absolute error, mean percent forecast error, root mean square error and Theil's inequality coefficients. IMA (0, 1, 1) model performed well for forecasting purposes. The percent prediction error for the last three years i.e. from 2015–16 and 2017–18, was below 3%. The predicted values along with their standard errors up to the year 2099, were also obtained using the model.
Akaike information criterion
Bayesian information criterion
Mean absolute error
Data set
Box–Jenkins
Cite
Citations (3)
Seasonality
Box–Jenkins
Mean absolute error
Cite
Citations (187)
This study conducts a Time Series Analysis of air quality indicators with an aim to forecast the values of air pollutant concentrations like SO2, NO2 and PM10, using Auto Regressive Integrated Moving Average (ARIMA) and Long Short Term Memory (LSTM), for the cities of Margao and Sanguem in the Indian state of Goa. On comparing the results from the two methods, based on the performance measures of Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) it is observed that LSTM model achieves a fair prediction of the air pollutant concentration values.
Mean absolute error
Moving average
Cite
Citations (0)
Akaike information criterion
Partial autocorrelation function
Bayesian information criterion
Moving average
Autoregressive–moving-average model
Box–Jenkins
Cite
Citations (210)
Akaike information criterion
Bayesian information criterion
Mean absolute error
Cite
Citations (8)
Abstract PM 10 is one of the aerosol particles that can endanger human health. This research conducted by forcasteing for PM 10 concentration. Forecasting is an activity of estimating or predicting events in the future, therefore it is necessary to do analysis simple statistical model to know goog results. In this case, several forecasting models are used for the daily PM 10 concentration in Kototabang, that is Liniear, Quadratic, and Exponential Trend Model. As the results of this research, monthly forecasting using Linear Trend Model has the highest correlation value in October (-60) with MAD value 0.254, MSE 0.0651, RMSE 0.255, and MAPE 1848.8. Monthly forecasting using the Quadratic Trend Model has the highest correlation value in February (+0.52) with MAD value 0.013, MSE 0.0002, RMSE 0.015, and MAPE 103.934. then, for monthly forecasting using the Exponential Trend Model has the highest value in October (+0.61) with MAD value 0.124, MSE 0.0154, RMSE 0.124, and MAPE 893.484. the output of the forecasting model obtained the best model for forecasting the daily PM 10 concentration in Kototabang i.e by using a simple statistical model, linear trend model.
Quadratic model
Value (mathematics)
Cite
Citations (1)