Multi-step sales forecasting in automotive industry based on structural relationship identification

Forecasting German car sales using Google data and multivariate models

“…All data were transformed into logarithms to reduce variability and convert nonlinear patterns to 4 60 80 100 120 140 160 180 200 01 02 03 04 05 06 07 08 09 10 11 12 13 14 BUILDING CONSTRUCTION ORDERS 60 70 80 90 100 110 120 01 02 03 04 05 06 07 08 09 10 11 12 13 14 CONSUMER CONFIDENCE INDEX 85 90 95 100 105 110 01 02 03 04 05 06 07 08 09 10 11 12 13 14 CPI 0 1 2 3 4 5 6 01 02 03 04 05 06 07 08 09 10 11 12 13 14 EURIBOR 500 550 600 650 700 750 01 02 03 04 05 06 07 08 09 10 11 12 13 14 GDP 70 80 90 100 110 120 130 140 01 02 03 04 05 06 07 08 09 10 11 12 13 14 PRODUCTION INDEX 6 7 8 9 10 11 12 13 01 02 03 04 05 06 07 08 09 10 11 12 13 14 UNEMPLOYMENT RATE 70 80 90 100 110 120 130 140 01 02 03 04 05 06 07 08 09 10 11 12 13 14 PETROL PRICE Figure 2: Economic variables -not seasonally adjusted. Sample: 2001M1 -2014M6 linear patterns 2 (see Sa-ngasoongsong, Bukkapatnam, Kim, Iyer, and Suresh (2012)). The descriptive statistics for the car registrations, the Google data and the economic variables (both seasonally adjusted and raw data) are not reported for the sake of space and are available from the authors upon request.…”

“…To select the best multivariate model for each car brand, we follow the structural relationship identification methodology discussed by Sa-ngasoongsong, Bukkapatnam, Kim, Iyer, and Suresh (2012) for the case of the US car market. Briefly, the first step is to identify the order of integration using unit root tests; if all variables are stationary, VAR and VARX (Vector Autoregressive with exogenous variables) models are used.…”

“…The third group includes various political, economic and social environmental factors which are generally beyond the control of manufacturers, such as organizational issues, political issues, global economic growth, ecological and physical forces, socio-cultural effects and consumer behavior. The use of these factors for car sales forecasting has been rather limited, see Brühl, Borscheid, Friedrich, and Reith (2009), Shahabuddin (2009), Wang, Chang, and Tzeng (2011) and Sa-ngasoongsong, Bukkapatnam, Kim, Iyer, and Suresh (2012). Moreover, most previous studies have focused on the dynamics of car sales in the short-term, with forecast horizons usually less than 4 months, whereas car sales forecasting requires time scales with duration up to one year or more.…”

Chinese automobile sales forecasting using economic indicators and typical domestic brand automobile sales data: A method based on econometric model

“…Finally, the two forecasting performances of considering the influence of Chery sales and ignoring it are considered. Dekimpe et al 6 and Sa-ngasoongsong et al 14 thought that VAR and VECM are powerful, which are theory-driven models that can be used to describe the long-run dynamic behavior of multivariate time series. In addition, VAR and VECM in the timeseries model are introduced in this article, and then, their forecasting performances are compared.…”

The Combination Forecasting Model of Auto Sales Based on Seasonal Index and RBF Neural Network

“…A new market forecasting system for tobacco wholesalers in China based on a new developed market demand forecasting model is presented, this model can help sales company to improve the forecasting accuracy for annual and monthly market demands significantly [4]. Nonlinear and non-stationary evolution and dependencies with diverse macroeconomic variables hinder accurate long-term prediction of the future of automotive sales, a structural relationship identification methodology is presented to identify the dynamic couplings among automobile sales and economic indicators, the empirical results suggest that VECM can significantly improve prediction accuracy of automotive sales for 12-month ahead prediction compared to the classical time series techniques [5].Using the combination forecasting method which based on PLS to predict the province's cigarette sales of the next year, the results show that the prediction accuracy is good, which could provide a certain reference to cigarette sales forecasting in a province [6]. Two strategies for forecasting -using the set of good forecasters in the testing phase versus filtering out the bad forecasters and using the resulting set of forecasters were compared, experimental results with over 30 sales series indicate that our heuristics greatly out-perform those such as the simple mean [7].…”