The Prediction Of Hotel Decoration Project Cost Based On Principal Component Regression Model

With the rapid development of China’s economy, national income also will increase, at the same time also led the development of the consumer market, people’s consumption concept has gradually changed. People living in the house decoration more beautiful, around four or five-star hotels have sprung up, we can see people for the requirements of the building is no longer a shelter can be, but it requires a more beautiful, more comfortable, this change will bring opportunities for architectural decoration industry.Where there are opportunities, there are challenges, architectural decoration market become more and more mature and it has brought enormous pressure to each decoration company, the competition among companies has become increasingly fierce, in this case, for each decoration companies, to do a scientific cost management will become more and more important. Project cost forecasting is an important part of project cost management, it is the most important part of the engineering project cost dynamic control, it is one of the determinants of corporate profits or not, and it also is the basis of decision making for companies bid price in the bidding stage, so, it is possible for the project to accurately predict the cost of enterprise should have become the core competitiveness. In this paper, from the perspective of decoration construction enterprises, to forecasting the cost of proposed project-H hotel decoration project, proposed using principal component analysis and linear regression method of combining the cost estimates for the proposed project. The content of the thesis are as follows:1. According to the proposed project has been forecast, collect a large of historical data and information from the similar completed projects, after read a lot of literature and consult with the expert project managers, screened out the factors which are impacting the hotel decoration project cost, extract the corresponding information about these factors from the completed projects to do a basic work for the conduct of cost prediction.2. According the extracted impact factors, the author proposed using principal component analysis to do a preliminary analysis and extracted a little principal component factors to express the most information the original factors has expressed, not only eliminate the existence of multicollinearity, but also reduce the amount of factors for the establishment of a linear regression model.3. By SPSS statistical software, finished the factor’s principal component analysis, use the new principal components factor to establish a principal component regression model, and do a significant test for the model, verified its feasibility through examples.4. Using the established principal component regression model to make a cost prediction of the proposed project-H hotel decoration project, compared the result of prediction and actual result, and analysis the reason why there have a deviation between the two results.