An Investment Analysis And Forecast Of A LED Company

In 2014, the output scale of LED industry in China is nearly 344.5 billion RMB, increasing 31%. And the output value of LED downstream applications is 276 billion RMB, increasing 32%, which has a broad market prospect. The company which is involved in this paper belongs to LED downstream application market. In 2014, there is a big gap between the expectations of the LED output grow amount and the real amount. In this case, this company needs to do predictive analysis of the business status. Compared the forecast results with peers, the company can find out the reasons and do some adjustments to the company’s industrial structure, achieving the company’s value growth and increasing the economic efficiency.In this paper, the operating income data from 2008 to 2014 in this company is collected. After sorting these data, timing chart is made; ARIMA model and Holt-Winters multiplicative models are established. About the predicted results and the actual results, it shows that the effect of Holt-Winters multiplicative model is much better than ARIMA model, but its prediction accuracy is still not ideal. So the prediction result of the Holt-Winters multiplicative model is processed as input of the BP neural network, and residual is processed as output, improving forecasting accuracy. These processes are all about Winters-BP prediction model which is applied in this article. Using Winters-BP combination model and four quarters of total operating income of the company in 2015, revenues of various domestic and foreign industries are judged, which can be used to provide appropriate reference for the company’s development strategy and help companies make informed decisions.