Search

When a forecast of the total value over several time periods ahead is required, forecasters are presented with two temporal aggregation (TA) approaches to produce required forecasts: i) aggregated forecast (AF) or ii) aggregate data using non-overlapping temporal aggregation (AD). Often, the recommendation is to aggregate data to a frequency relevant to the decision the eventual forecast will support and then produce the forecast. However, this might not be always the best choice and we argue that both AF and AD approaches may outperform each other in different situations. Moreover, there is a lack of evidence on what indicators may determine the superiority of each approach. We design and execute an empirical experiment framework to first explore the performance of these approaches using monthly time series of M4 competition dataset. We further turn the problem into a classification supervised learning by constructing a database consisting of features of each time series as predictor and model class labelled as AF/AD as response/outcome. We then build machine learning algorithms to investigate the association between time series features and the performance of AF and AD. Our findings suggest that both AF and AD approaches may not consistently generate accurate results for every individual series. AF is shown to be significantly better than AD for the monthly M4 time series, especially for longer horizons. We build several machine learning approaches using a set of extracted time series features as input to predict accurately whether AD or AF should be used. We find out that Random Forest (RF) is the most accurate approach in correctly classifying the outcome assessed both by statistical measures such as misclassification error, F-statistics, area under the curve, and a utility measure. The RF approach reveals that curvature, nonlinearity, seas_pacf, unitroot_pp, mean, ARCHM.LM, Coefficient of Variation, stability, linearity, and max_level_shif are among the most important features in driving the predictions of the model. Our findings indicate that the strength of trend, ARCH.LM, hurst, autocorrelation lag 1, unitroot_pp, and seas_pacf may favour AF approach, while lumpiness, entropy, nonlinearity, curvature, and strength of seasonality may increase the chance of AD performing better. We conclude the study by summarising the findings and present an agenda for further research. [ABSTRACT FROM AUTHOR]

People typically spend most of their time indoors. It is of importance to establish prediction models to estimate PM 2.5 concentration in indoor environments (e.g., residential households) to allow accurate assessments of exposure in epidemiological studies. This study aimed to develop models to predict PM 2.5 concentration in residential households. PM 2.5 concentration and related parameters (e.g., basic information about the households and ventilation settings) were collected in 116 households during the winter and summer seasons in Hong Kong. Outdoor PM 2.5 concentration at households was estimated using a land-use regression model. The random forest machine learning algorithm was then applied to develop indoor PM 2.5 prediction models. The results show that the random forest model achieved a promising predictive accuracy, with R2 and cross-validation R2 values of 0.93 and 0.65, respectively. Outdoor PM 2.5 concentration was the most important predictor variable, followed in descending order by the household marked number, outdoor temperature, outdoor relative humidity, average household area and air conditioning. The external validation result using an independent dataset confirmed the potential application of the random forest model, with an R2 value of 0.47. Overall, this study shows the value of a combined land-use regression and machine learning approach in establishing indoor PM 2.5 prediction models that provide a relatively accurate assessment of exposure for use in epidemiological studies. Image 1 • A practical indoor PM 2.5 prediction framework was proposed. • The combined land-use regression and machine learning approach was applied. • External validation was proposed to evaluate the prediction model. • The random forest model outperformed the linear mixed-effect regression model. • The outdoor PM 2.5 concentration is the most important predictor variable. [ABSTRACT FROM AUTHOR]